Taboola.com Ltd Aktie

Good day, and thank you for standing by. Welcome to the Taboola Third Quarter 2026 Earnings Conference Call. [Operator Instructions] Please be advised that today's conference is being recorded.

I would now like to hand the conference over to your first speaker today, Aadam Anwar, Head of Investor Relations. Please go ahead.

Thank you, and good morning, everyone. Welcome to Taboola's Third Quarter 2026 Earnings Conference Call. I'm here with Adam Singolda, Taboola's Founder and CEO; and Steve Walker, Taboola's CFO. The company issued earnings materials today before market opened, and they're available in the Investors section of Taboola's website. Now I'll quickly cover the safe harbor. Certain statements today, including our expectations for future periods, are forward-looking statements. They are not facts and are subject to material risks and uncertainties described in our SEC filings. These statements are based on currently available information, and we undertake no duty to update them, except as required by law. Today's discussion is also subject to forward-looking statement limitations in the earnings press release. Future events could differ materially and adversely from those anticipated.

During this call, we will use terms defined in the earnings release and refer to non-GAAP financial measures. For definitions and reconciliations to GAAP, please refer to the non-GAAP tables in the earnings press release posted on our website.

With that, I'll turn the call over to Adam.

Thanks, Aadam. Good morning, everyone, and thank you for joining us today. In the third quarter, we delivered $189 million in ex-TAC, growing 7% year-over-year and $56 million in adjusted EBITDA, growing 16%. The environment in which we operate is changing. We're seeing shifts in how consumers discover and engage with content across the open web, including the growing influence of AI summaries and LLM. While page views across our publisher base remain relatively steady, we are seeing advertising impressions decline as consumer behavior changes. We believe our publishers had meaningful opportunities to use our products to improve engagement and grow revenue, and I'll discuss those opportunities in more detail later.

In the near term, however, we're incorporating these trends into our outlook and lowering our full year ex-TAC guidance to $758 million, representing 6% growth. To address these dynamics, we're doing 3 things as part of our realized strategy, which is centered around helping advertisers find outcomes at scale beyond search and social. First, we're focused on winning new publisher partnerships. This is evidenced by our recent signing of Fox News, Fox Local, Fox Sports and Fox Weather, which are premium properties that advertisers love.

We have a strong pipeline of additional publishers around the world that we hope to bring on to our platform, creating more opportunities to drive engagement for publishers, revenue for our partners and performance for advertisers. Second, we're focused on growing the revenue we generate with each publisher. Historically, we've monetized mainly through native placements. However, with products like full page monetization, we monetize the entire ad inventory of a publisher, including display, vertical formats, native and more. We believe we can increase that monetization 2 to 3x per publisher. So this is something we're leaning into with our teams and our top publishers.

As an update on full-fledged monetization, last quarter, we announced NBCUniversal as our first full-fledged monetization partner. We're now live across 100% of the advertising inventory on NBC News and today.com, and the performance we're seeing is on par with what NBC and us expected, which is encouraging. For publishers, this creates an opportunity to grow revenue, while working with fewer partners, reducing operational complexity and improving efficiency. For Taboola, it means growing our share of wallet with existing publishers while giving advertisers access to even more premium inventory. Another way we're growing revenue for our existing publishers is through DeeperDive, our AI answer engine, which is now being used by nearly 14 million people across publishers and OEMs, which gives publishers valuable insight into what their audiences care about.

As consumers increasingly turn into AI to discover information, DeeperDive positions Taboola to help publishers adopt while creating another way to grow revenue from our existing relationships. We're also expanding our publishers' relationships and the revenue we can generate with them through our proposed acquisition of Dianomi. This is an opportunity to broaden our publisher base and further strengthen our position in business, finance and lifestyle verticals, areas we already have strong presence and where advertisers value access to high-intent audiences. Dianomi brings a highly specialized network of hundreds of premium publishers, including Reuters, CNN Business and the Wall Street Journal, alongside blue-chip financial advertisers.

By combining our companies, we believe there's a monetization growth opportunity across Dianomi premium publishers driven by realized AI and advertisers as well as Taboola given Dianomi expertise in high-intent verticals and premium advertisers. We expect to complete the acquisition before the end of 2026, and we will continue to provide updates. These are all examples of ways we're doing more with our existing publishers to grow revenue.

Thirdly, we're investing in diversified areas of our business that are growing and not impacted by AI, such as Taboola News, apps and social commerce. These markets give us access to new audiences while allowing us to leverage the assets we've built over the last decade, including our technology, unique first-party data, publishers relationships and global distribution. Our goal is to build multiple growth engines that complement our core business and increase Taboola's competitive advantage across the broader digital ecosystem. In times of change, I'm convinced that publishers and advertisers now more than ever before, would want to innovate, move even faster and look for revenue-generating opportunities. Much like in social and search, the OpenWeb is the opportunity to have one company that can be the monetization infrastructure for everything, and I believe it can be Taboola.

To wrap up, our focus is on our partners and clients working with them closely on using our wide array of products offering to grow revenue, especially during times of change in our industry. We were never more aligned than we are now.

And with that, I'll hand it over to Steve.

Thanks, Adam, and good morning, everyone. Q3 results were in line with our guidance. As Adam mentioned, the environment in which we operate continues to change and the way consumers interact with our partners' web properties is being affected by the influence of AI and LLMs. While page views remain relatively steady, what we have seen has been a reduction in the number of impressions available for our advertisers and a headwind on revenue. As a result, we are adjusting our Q4 guidance, which I will share in more detail shortly. But first, let me talk about our Q3 results.

In the third quarter, revenues were $462.3 million. Our revenue was primarily impacted by the impression volume headwinds and by a slower-than-expected rollout of Next Engage, which is the product we introduced to replace ExploreMore. When looking at these numbers on a year-over-year basis, revenue was down 7%. Last quarter, we discussed our efforts to optimize the quality of our supply and the deprecation of the ExploreMore product. These factors, along with some headwinds from the decline in impressions due to consumer behavior changes drove the decline.

In Q3, the number of scaled advertisers declined 4% year-over-year and average revenue per scaled advertiser declined 5%. Scaled advertisers represent approximately 85% of revenues. So when we see declining revenue as we did this quarter, we would expect to see a decline in these metrics. Ex-TAC gross profit increased 7% year-over-year to $188.8 million, and gross profit for the quarter was $152.3 million, up 10% year-over-year. Ex-TAC gross profit benefited from a couple of factors. First, as the volume of our supply declines, we see higher ad rates on our network because a similar amount of demand is spread over less supply. Higher ad rates drive higher ex-TAC margins.

Second, we saw strong growth in Taboola News, which contributed to the growth in ex-TAC. Net income for the quarter was $18.4 million with non-GAAP net income coming in at $39.9 million. Adjusted EBITDA for the quarter was $56 million, which represented a margin of 29.7%. This reflects our ongoing discipline in expense management as operating cash expenses remained broadly stable, excluding foreign exchange impacts. Foreign exchange continues to be a headwind in 2026 on a year-over-year basis.

On a constant currency basis, FX represented a $3.4 million headwind to third quarter adjusted EBITDA. Excluding this impact, adjusted EBITDA would have been approximately $59.8 million, which would have represented an adjusted EBITDA margin of 31.6%. We expect FX to remain a headwind in Q4. In terms of cash generation, we had $35.6 million in operating cash flow in the third quarter and free cash flow of $20.4 million. We expect our free cash flow to convert from adjusted EBITDA at a 60% to 70% rate this year.

Turning to the balance sheet. We remain in a strong financial position. We ended the third quarter with a net cash balance of $50.2 million. Cash and cash equivalents totaled $128.7 million, which more than offset our long-term debt of $78.5 million. As of September 30, we had approximately $191 million of available liquidity under our $270 million revolving credit facility. In the third quarter, we repurchased approximately 7 million shares at an average price of $4.37 for a total consideration of $30.3 million. We have approximately $90 million remaining under our authorization.

We also received an additional $200 million in authorization for share repurchases. This provides us with flexibility to return capital to shareholders when we believe it is appropriate. As you've seen this quarter, we continuously evaluate capital allocation opportunities, including share repurchases, investments in our business and strategic acquisitions such as Dianomi with the goal of allocating capital where we believe it can generate the greatest long-term value.

Moving to guidance. For the fourth quarter, we expect revenues to be between $477 million and $493 million, gross profit to be between $174 million and $183 million, ex-TAC gross profit to be $204.5 million to $213.5 million, adjusted EBITDA to range from $73.5 million to $82.5 million and non-GAAP GAAP net income to be $53 million to $62 million.

For the full year, we are updating our guidance as well. We now expect revenues to be between $1.88 billion and $1.9 billion, gross profit to be between $595.5 million and $604.5 million, ex-TAC gross profit to be $754 million to $763 million, adjusted EBITDA to range from $212 million to $221 million and non-GAAP net income to be $151.5 million to $160.5 million. In summary, we're seeing changes in consumer behavior, but are working to offset these changes by bringing on new publishers organically, such as with FOX News and via acquisition, such as through the proposed acquisition of Dianomi and by doing more with our existing publishers through full page monetization.

While we're still in the early stages of many of these initiatives, we're focused on executing, building on our competitive advantages and delivering long-term growth and value for our shareholders.

With that, let's move to Q&A. Operator, can you please open the line for questions?

[Operator Instructions] Our first question comes from Naved Khan at B. Riley Securities.

Just a couple of questions for me. Maybe first on the dynamics you just described, Adam, about page views being stable, but impressions being lower. Is that just a factor of sort of visitors to the site scrolling less and looking more at summaries? Or is there something deeper than that, that's going on? So that's part one of my question. The second part is about the DeeperDive. Are you seeing more distribution across more publishers for this offering? Just talk about that and also how the monetization is progressing on DeeperDive.

Yes. Steve, do you want to take the pages?

Sure. Yes, I can start with what Adam was talking about on the page view and impression side of things. So what we're seeing is basically what we think is a shift in consumer behavior. It created a headwind on our impression volume. So while page views are relatively stable, we're seeing a decline in -- or fewer impressions per page view. We think this probably has to do with consumer behavior, maybe not scrolling as much or something along those lines. But generally speaking that, for us, the important thing is that it's fewer impressions, fewer ad impressions that we have to fill. And that's why we're now focused on, as we've talked about in the past and what Adam talked about, winning more new publishers organically and through acquisition like with Dianomi, doing more with our pubs. So explore more -- sorry, full page monetization, DeeperDive, that all gives us more impressions and then also investing in areas that are less impacted by those consumer shifts. So for instance, Taboola News and social commerce are 2 areas we're leaning into. So generally speaking, that's what we're seeing and kind of what we're doing about it.

And then on the second question on DeeperDive. So if this was a startup, it would be a very, very fast-growing start-up. It's still early in terms of revenue impact for Taboola given our size. But what we're seeing is essentially a very accelerated growth of publishers integrating DeeperDive. It's been around for about a year now or so, and we just crossed a few interesting milestones. More than 100 sites are using DeeperDive. More than 14 million people are interacting with DeeperDive which is fast growing. And I have no doubt that as I look into kind of the future, any company, even right now, Motorola, which we announced a few weeks ago, which is a device manufacturer, a big one, has integrated DeeperDive. So what you're seeing is that publishers, device manufacturers, I think over time, apps, even TV, I suspect any kind of digital canvas will offer a conversation on there. And there's an opportunity for us here to bring kind of our AI expertise, first-party data and experience with DeeperDive to all of those touch points. So I just think that's going to be a growing trend over the next 2, 3 years that consumers will expect it. And that's also going to help alleviate some of the, I think, the ad impression pressure that we're seeing with more touch points with consumers. So the answer is that we're seeing good growth, a very fast one. If this was a startup, it would be one of the fastest growing startup.

Our next question comes from Laura Martin at Needham.

So let's start with traffic. The guidance you gave for the fourth quarter is coming down a lot from our prior estimates. And Reddit had said on their call that Google had made another change in its search algorithms and that traffic had gotten much worse in October. Is that the cause of bringing the guidance down in the fourth quarter for you guys as well as Google search monetization and referral traffic shifts? And then my other question, Adam, is, could you give us an update on how you guys are using AI in the business to drive business results? Because I know you had done an agentic tie-in with Anthropic, and I'm interested in what's happening with some of your AI products and what kind of luck you've had in driving monetization upside from AI.

Thanks, Laura. So I'll answer the first one. So the simple answer to your question is, no, it's not less referral traffic from Google that we're seeing. It's really about the fewer impressions per page view. So it's really something about consumer behavior when they're on our pages. So it's not fewer page views. It's less engagement once they're on the page or at least seeing fewer of our ads.

Yes. So a few things. There's a lot at I think at Taboola that we're focusing on as it relates to AI. Internally, our President and COO is getting a big project of how to empower our people to do even more as it relates to our clients and internally and drive kind of productivity and revenue through that. So that's a big one for us. And then externally, DeeperDive, obviously, is a big AI initiative for us. We're -- I think we're one of the only companies that I'm aware of that actually is showing ads in an LLM conversational layer. And we did mention that we're seeing very good results from those ads. We announced a quarter plus ago our integration with Cloud, which essentially allows through an MCP with Cloud any kind of company, anyone to use their own cloud on their computer and launch a campaign with Taboola. I just published a video that I'm interviewing one of our VPs about that.

And I think the biggest -- to me, the biggest area where I see kind of revenue opportunity about that, Laura, is with agencies. I think agencies, the big holdcos are spending a lot of time with us at a very senior kind of level, imagine in a new version of themselves, whereby they don't have to go through the traditional ad tech. And with us, they can build agents, agentic agents that can reach out directly to Taboola and negotiate directly with us and our publishers and distribution based on whatever the client needs. So I do think that the way it's going is that big clients will basically have agents that will negotiate with suppliers directly as it relates to their own goal, whether that's subscription or purchases or whatever they may need. And by doing so, we'll bypass the whole kind of ecosystem. And that's maybe where the biggest kind of upside from a revenue perspective. And it takes good attention from me, Krishan, our Chief Business Officer and the team as we're looking into the future with kind of advertisers.

Our next question comes from Tyler DiMatteo at BTIG.

I wanted to come back to the impressions and the page views here. I understand page views have been steady. But I guess at the highest level guys, when did you start to see that those impressions come down? I guess, like what's the visibility into that? And then has that accelerated? Obviously, you're reflecting it in the 4Q guide. But I'm just curious, like we're sitting here in November, like what does that look like? And when did you start to see those impressions go down? And then secondarily, my next question here, operationally, when you just kind of take a step back here and you're looking at all the different products and things that you could do in adding incremental premium publishers, what do you feel you have the most control of now in this environment?

Yes. So let me -- I'll start with the first part of your question. So I think in terms of timing of this, we saw signs of it earlier this year, but it was frankly overshadowed by the Explore More deprecation and some of the network cleanup that we were doing. So it wasn't as noticeable. And frankly, it was being offset mostly by higher ad rates that we were able to achieve. So it didn't impact revenue as much. It became pretty clear in Q3, which is obviously why we're talking about it now because we had a relatively clean quarter. Explore More obviously, was in our past. And so we were able to see this much more clearly. So while we saw indications earlier, Q3 is when we really saw it. In terms of thinking about going forward on this, it's obviously baked into our Q4 guide. So we've seen it. We recognize that we've baked it in. I think, though, looking to next year, we'll obviously talk about that when we get to February and going forward.

I think to your second part of your question about what do we really have control of, I think Adam talked about this in his prepared remarks. We have control of winning new publishers. So we're out working on controlling or winning new publishers. We're also obviously acquiring a really good high-quality publisher base by acquiring Dianomi. We also can control doing more with new publishers. So that's why we're out there selling full page monetization and working with more publishers on expanding what we do, selling them DeeperDive and doing more there. And we can obviously also control investment areas in areas that are less impacted by AI. So Taboola News, social commerce.

So I think those are the things that we feel are in our control. We obviously can't control what consumers are going to do day-to-day, but we still can grow our supply by leaning into all those other things.

Right. And I do think that now more than ever, when you talk to publishers and spend time with them as well as advertisers, everyone is looking for revenue growth, these times of change are also -- as I've seen in the past also with Taboola, it's always the biggest opportunity to talk to your partners, deepening the relationship, spend time about the future and to be able to be in a position where publishers can hand to Taboola the entire ad inventory with full page monetization as an example, is a big deal. So we'll continue to update. I'm personally involved with many of our bigger publishers kind of talking to us about it into the future, and it's quite exciting. We'll see how that progresses. But again, I look at this period in a period of people want to do a lot more with the partners they trust in the relationship with.

Our next question comes from Barton Crockett at Rosenblatt.

I wanted to maybe see if you could explain why we have this change. So I understand that people are apparently seeing fewer impressions, which I guess suggests they're not spending as much time browsing or reading the open web. But why? Why would that be happening? I think is this in any way tied to AI overviews -- because I know you guys have said in the past, your AI overview exposure is single digits. So some clarity would be great.

Yes. It's a good question. And it is -- we're very fortunate that most of our publishers are big, which means they have a lot of direct traffic and our search exposure is fairly small. And that's why the page view kind of change is fairly small. It's fairly steady. I do think though that consumers now that come to our publisher site behave differently. They expect things to be at the top, maybe a summary in the article or in the home page. And it's hard to really narrow down the reverse engineer exactly why they do that. But I think the traffic that comes to the publishers versus the traffic that stays on Google when they see the summary is affecting kind of the type of ad impressions that we're seeing. And I think it's a variety of scrolling, it's a variety of the type of articles they come to read versus stay on Google, things of that nature. And I also suspect this will plateau at some point because it's going to be mature enough at some point. But right now, we have seen this change, and that's why we wanted to put this in the guidance for Q4. But I do think to your question, it's a combination of the type of articles people do read versus stay on the search page as well as kind of expectation to get things at the top of the page versus scrolling to the bottom. And that's also why DeeperDive, which is at the top of the page as well as full page monetization, which is all inventory matters to us because we basically want to see consumers wherever they may be on our publisher site. And like I said, I'm encouraged by the fact that people really want to be in relationship with the big publishers that we work with, and that's why the pages are steady, but there is a change in behavior.

And then could you talk a little bit about the breadth? I mean, is this just a few particular publishers of a particular kind of nature or is it everyone? Are you seeing it on Taboola clients and people who aren't Taboola clients? And do you have any documentation to kind of support your view of the breadth of this?

So first of all, it is kind of across the board. I think some a bit more, some a bit less. Again, most of our publishers are similar in the sense that their search traffic is fairly small and their direct traffic is fairly high. So we don't have a lot of kind of like very high search traffic sites that probably would have seen different behavior. So it kind of impacts everyone to some degree, some a bit less, some a bit more. And I don't know if we have something that we publish also, we don't -- we want to be kind of sensitive to our partners and not share things that we're not want to share, but we can look into see if there's any insights we can share on this.

Maybe we can do a block post or something share what we're seeing.

Okay. And then if I could just ask one other thing, just to clarify, the implied kind of year-over-year trajectory in the fourth quarter guide, is there any reason why we shouldn't presume that, that's a base case for '27?

Yes. So we don't want to obviously guide for 2027 right now. But I think, obviously, we updated our Q4 guide to reflect what we're seeing. So I think you can use it as an indication. We'll obviously provide more guidance in February. In the meantime, I think our ambition as a management team is to get back to double-digit growth, and that's where our focus is.

Our next question comes from Alex Lavigne at StoneX.

Just on Realize+, curious what percentage of advertiser spend is leveraging Realize+ today and what the incremental ROAS compared to legacy campaigns has been? And then related, how meaningful is the spend coming through the Quad MCP integration? And is this largely new budget or existing, acknowledging you mentioned Holdco as an opportunity?

Yes. So on the Realize+, I don't have those figures in front of me, to be honest with you. I know that the adoption rate on Realize+ have been significant, and it was one of our fastest adopted products in our release history. So I know it was being adopted by a significant number. I don't have the exact number in front of me. I think we talked last quarter about it being over 1,000 advertisers utilizing it, but I don't have an updated number on that.

You want to talk about the MCP?

Yes, the MCP is still -- it's millions of dollars, but -- so fairly small. But I would say with 0 investment or any promotion of that, this is money that's kind of like just coming our way, people experimenting with it. So again, I think the biggest growth on that one, my suspicion will be from agencies and very large clients who will see an opportunity to kind of collapse ad tech and buy directly through an agent from suppliers, not only Taboola, by the way, I expect this will be the behavior across the board. They will build agents that will negotiate with big suppliers on search, social and beyond search and social, where Taboola is obviously big. So we're worth negotiating with. So that's the biggest upside from that kind of initiative I'm seeing with Cloud, at least in my opinion.

This concludes the question-and-answer session. I would now like to turn it back to Adam Singolda for closing remarks.

Thanks, everyone, for joining us this morning. So as I said, the AI and reshape how consumers discover content, we're focused on 3 main priorities. The first one is expanding our publisher base. You've seen wins like Fox News, and there's a very strong pipeline ahead of us, and we'll continue to update, and that's going to help us grow more supply and more impressions, which is a focus of us. The second one is deepening our relationship with existing publishers through initiatives like full page monetization, DeeperDive, newsroom and many other things that can grow engagement and increase supply we have on our existing publishers. And the third one is investing in diversified areas of our business where AI is not impacting those, things like Taboola News and e-commerce where we're seeing faster growth than the rest of the business. I'm convinced that now more than ever, the OpenWeb and everything outside of search and social deserves a platform that can drive revenue through advertising and through AI initiatives, and Taboola can be that company. Thank you for the support, and we look forward to speaking with all of you over the next few weeks.

Thank you for your participation in today's conference. This does conclude the program. You may now disconnect.

Perfect. So we'll get started here. Adam, welcome. Let's get started. My name is Ron Josey, I cover the Internet sector here at Citi. And I'm always excited to have my old friend, Adam Singolda here with us today, the CEO, Founder, Co-Founder of Taboola. We've known each other for a number of years now, I think. I was thinking about like what the right icebreaker is, when we were talking about this just now as we catch up. Talking about your latest LEGO set and then we get into like the real business.

I'm building the Formula One car. It's probably like 2,000 pieces and not that big. I was trying to count.

With the kids or on your own?

That's mine. No, so we have -- each have their own. My son is doing this like art piece, my daughter is doing a unicorn and other one is doing, I think, a Super Mario. So we have a -- I realize that I enjoy, I matter too. I can't just do things with them all the time. So that's where I'm right now. It's fun.

You matter, too. I think that's -- I don't disagree. So why don't we kick off here? And I would love to level set for everybody in the audience listening on the webcast. Just give us an overview of Taboola just the evolution of performance advertising has been pretty extraordinary. It just continues to be a greater part of most companies' budgets. And so just talk to us about Taboola's approach and where things are going from a performance advertising perspective.

One, I think over the last few years, you've seen -- for those who may not know, Taboola is the largest kind of outcome or performance advertising platform outside of Google and Facebook. So we specialize in helping businesses big and small to reach a certain outcome that matches their growth strategy. So they always go to Google first. They go to Meta, then they get kept out and they're looking for other partners that can help them satisfy that. The definition of what's outside of the way search and social changed over the years. It used to be publishers at WWW, someone.

And then over the last, really, 5, 10 years, it evolves into anyone that wants a piece of the advertising market. Uber is making money from ads. The #1 line of EBITDA for Amazon is ads and services. Netflix said they'll never do ads, now they love ads. Everyone is in the advertising business, and everyone needs a good friend that's not maybe Google or Facebook. And that's where we fit in. So I think the market has evolved to your question from just traditional open web websites to apps, OEM manufacturers, utility, e-mail providers, just anyone that wants to make money from ads. That's one thing to change.

Two, the world is crazy. So the idea that you can spend money not knowing what's going to happen with that money over time went down. You really expected to have an attribution model that allows you to spend and know that it is working. And so that's also changed. And the third minute change is AI, like that's reshuffled everything, conversational ads, the idea that consumers now talk to you, how is that going to be monetized, what type of data that creates. We just had on stage here before we do understand we had Microsoft today, and that's obviously a big part of what they talked about.

So let's get into that in terms of how the online ad market is changing. Before I do, if you could just give us a level set of the online ad market in your view, healthy, building, challenged? How would you look at the market overall in terms of demand?

I think it's, for the most part, stable. I think especially as there's more focus on performance. Performance advertising tends to be more stable is the last thing you stop. So I would say, it's mostly stable. Of course, we're seeing pockets and geographies where things are a bit different. And then the biggest thing in the future, which I'm sure we'll get into is if consumers are going to be conversing so much with LLM, how is that going to transpire into like what does that mean to the economy and advertising. But outside of that, I think today is pretty stable.

Yes. Certain geos, we hear a lot about like decent in the U.S. from an advertising demand perspective. So yes. So with that, Adam, you actually -- you highlighted a topic that I think is coming up in every single conversation, which is how GenAI is changing how we all interacted with the web and agents you can easily see how you can have that -- you're already having a conversation with agents. A lot of companies are launching these agents that are your best friends that can do everything for you on your behalf. Just talk to us about how the Internet can evolve or how it evolves with agents being your front door to a certain extent, like things are changing in terms of how we go online and what we do online.

I would divide it into 2 parts that I think are interesting to understand as investors as well that are looking to spend money in this ecosystem or not. So on one side, you have the demand side. How will AI affect billions of dollars kind of flowing into the open web and anywhere else. And I think the biggest change there is the evolution from programmatic, which is a protocol that has been used for the last 30 years by agencies and advertisers to buy ad and you know this really well, you covered many companies in that space.

I'm smiling because 30 years seems like a long time, yet, we're still in the evolution of programmatic.

Well, I think it's going to evolve into nothing. I think it's evolving into -- I think it may be replaced -- over time, we can discuss less over time. But when you're an agency, and you're spending $60 billion a year on programmatic and ads, and you have an opportunity to build your own buying machine using Claude Code or Codex or maybe SpaceX with personnel and you can own the IP that buys those ads and you can bypass everything on the way straight to the supplier. You can go straight to the consumer and negotiate with a different agent. That is a massive revolution we have never seen in, AdTech in a very long time.

The idea that you don't need any more DSPs. You can just bypass everything goes straight to the consumer on the other side and negotiate with an agent and ask that agent questions to buy and place an ad. So I think that's one of the biggest changes that might happen to -- on the demand side, which will affect agencies, DSPs, AdTech in general, agents buying from agents versus programmatic protocol that is like an old knee like such an old protocol. It's so inefficient. It can take advantage of all the first part of data that's available out there. I think it's going to be deprecated. How fast, I don't know, but I think faster than we can imagine. So that's one change.

Maybe before with that change, agents buying from agents, we're hearing that more and more. That's on the, call it, B2B programmatic side, I guess, is the question. Is that happening today? And give us an example of what that could be.

So for us, we announced a quarter ago that we've launched a Claude skill and an MCP, a bridge from anyone who wants to buy Taboola and place ads on Apple News, on ESPN, on Yahoo Mail, like amazing parts of the Internet through Claude Code, never talking to Taboola no need to know our dashboard, no need to talk to an account manager. No need to rely on our reporting everything through your Mac, like open your $20 subscription, open an account in Taboola and tell the Claude, place my ads on, I want sports as we know, is a big category. I want to be on ESPN. Can I do it? And you buy from Taboola.

We just made it available and it's already millions of dollars for us. Some people we don't even know. And I think so we're seeing signs of that. So I think it's happening in a big scale, No. But I can tell you all the holding companies, the agencies who spend tens of billions of dollars. That's all they do now. They're all building their agents buying arms that can interact with different suppliers that have unique data and unique distribution. I just don't know how fast it will happen. But look, we're seeing mergers between different agencies, we're seeing acquisitions for data purposes.

We're seeing actually things are happening. So I do think it will happen, and we're seeing small signs of that as of now, that I mentioned.

And these are -- I mean -- so with the skill that you built, so I've got 2 follow-up questions here. How long did it take you to build it, one. And then two, we're talking performance advertising, performance marketing. So this expands your funnel, your number of advertisers that could use Taboola to reach everybody within Yahoo Mail that meets a certain parameter. So talk to us about how you're advertising, what type of advertisers are coming on with this new skill. So how long does it take using AI in the development cycle to give us a sense? And then who's...

The time to build it was not that long because it still eventually uses Taboola's engine in the back end. So Claude is just talking. Claude became the best account manager, right? It's like 1,000 -- it's the best account managers we've ever had, and they're just using our system. So it's -- but it's still our engine called Realize. So Realize is our PMax or our kind of ad engine. So Claude is just the best account manager using that. So that's obviously still most of the work is on our side. But Claude just studied all of our documentation everything that was ever written and then just built all the directories it needs to come an expert.

So that wasn't that long. The type of advertisers is still roughly the same. So we're still very good with like mid-funnel kind of mid-sized advertisers. We're making a push towards bigger advertisers towards agencies because we think that's the right time to do that. And I do think still that most of the money that will come through MCP will still come from bigger advertisers and agencies, I think, for a while. For a time, I do want a lot of advertisers, but I don't suspect that will be most of our growth at least for the foreseen future. We'll have more advertisers but sizable ones. We call them scaled advertisers that essentially means over $100,000 a year, you become someone that has found a product market fit with Taboola.

That's helpful. That was one aspect of AI in the front door. What was the other aspect?

You have very good memory, Ron. The second thing is on the consumer side and the supply side. And I think over there, we're going to see a big part of the Internet as we know it transforming into conversational. So I mentioned USA TODAY, we've launched with them in September of last year, Deeper Dive, which essentially allows anyone to go to USA TODAY and ask any question they want about something that is covered by trusted humans that have covered the space. So the Knicks obviously won the championship. I hope you all love the Knicks and otherwise, you should leave the room immediately. But for the sake...

We have several Villanova graduates here. We thought...

That's great. It's my son's dream. So when that happened, and I went to USA TODAY, and I talked about the Knicks all the time and I find myself speaking 8, 9, 10 minutes about what the chances we're going to win compared this to a few years before. And I'm doing this no hallucination, just using USA TODAY content, I mean -- and then it says, by the way, if you're on that topic, read that article, you should read that and watch that. And it's a way better experience for me because when you love sports if you love sport you're never going to ChatGPT, hey, tell me yesterday's score. That's because it's not what you want an emotion. You want to watch the highlights. You want to get a point of view of the post maybe New York Post, you're part of the community. So to have that experience in the community and talk and watch and interact is such a better experience.

So I think on the supply side, we'll see a huge surge in different types of companies, making conversation available for their consumers. Look, the Lakers were bought for $12 billion. People believe that communities matter in the world of machines. So I do think that we'll see Deeper Dive, which is like services. I'm not saying we'll be the only one, but kind of like AI layers that offer consumers to do what they are now being trained to do, which is Talk to me.

How do you think we access the Talk to me? So who are -- is it going to be from one of ChatGPT's latest model? Is it going to be Meta's Hatch that we're talking about? Or is it going to be these newer bots that are coming on.

I can tell you how we do it. So we had a bunch of -- when we started that, we've had a bunch of open source models, and then we used our unique data about the consumer readership on the Internet to train those models and fine-tune them and refine them. And then we are lucky because we have a huge distribution of servers around the world. So from a CapEx perspective, where I think a lot of companies struggle because it's very expensive, and the APIs are very expensive. We were very fortunate we've had the CapEx to support hosting those models on our own servers and training the models based on our own unique data, which made it very affordable for us to make it available for free.

And the second thing is that how do you make money from that? Because otherwise, AI can just be a cost center. So we show -- again, we're lucky because most of our $2 billion in revenue is direct demand. So we could have sourced the right ad to monetize a conversation. And I can tell you and I spoke about this publicly, when you show an ad that's relevant for the consumer inside a conversation, the CPM of that ad and the value to the advertiser of that ad is 5 to 10x higher than an ad on the open web, a display on a home page. It is ridiculously valuable for the advertiser. The ROAS is so much better.

Because it's super targeted. And it's exactly what someone is having this conversation.

And I think you -- I think the consumers are having conversations to become super humans. They're like so intentional about what they want to do next. So it's like you prequalify those consumers. You are having 10 signals for the first time. We know Google is a good business. But for the first time, now USA TODAY to get a taste of Google Search business through Deeper Dive because those signals become available to them. And on USA TODAY, I mean it's more than 10 million questions a month. Can you imagine if you come to work and you get access to tens of millions of questions of your consumers on your own site.

Now you can decide what to cover from a news perspective, you can show ads that are relevant. So I think that if you can imagine, 20%, 30% of the open web evolved maybe 50% evolving to conversational type of experience, the revenue upside is huge. So much more money can be made and advertisers will find so much value in the open web. So I think that's exciting. And we're -- we've always been very close with publishers. So we're lucky that we get to have these conversations, and it's now live on many great partners, Advanced Media, Sinclair, Huffington Post, many great partners globally.

They are leveraging Deeper Dive on their site. So talk to us about Deeper Dive specifically because I think there's a lot of discussion out there in terms of all of my conversations might go through let's call it Gemini. Or might go through ChatGPT or might go through whatever Meta is doing with Hatch. But with Deeper Dive, if I understand it's correct, well, help me understand a Deeper Dive and b, then everyone become -- then all these publishers can be that engine on conversation. But tell me more about Deeper Dive.

So Deeper Dive is the first in the world free LLM answer engine supported by ads for the open web. You can be anyone if you work with Taboola, you can access Deeper Dive to make an answer engine, Gemini like on your homepage to go to usatoday.com. On their homepage there is ask us -- ask USA TODAY anything. We're using our first-party data to also generate questions for you. So a lot of consumers don't even know what that answer engine means. So when you go to USA TODAY, if you're a Knicks fan because you've been to ESPN, we may say -- a question will say, the Knicks season is about to start in October. Can we win again. And if I click on it, it's like, that's so nice, you know me. And if I click on it, I enter Deeper Dive on USA TODAY. It looks like a Gemini entered AI mode. You get the short answer. You can say I like it or I didn't like it, you can follow up and then Deeper Dive says, by the way while I have you, check out this video on USA TODAY about the Knicks. Checkout this article, look at this points of view about why we won and why we may win again.

And then I read it -- and then the Deeper Dive, again, this is not a Deeper Dive's article, how about this question, or maybe you want to talk to me more. And that is a completely new experience. It's free, and it's generating revenue every single time someone is using that. The adoption rates and you know the web really well. I mean, if you look at companies over the last 20 years, click-through rates on things online, like what is the likelihood I will click on an ad, what's the likelihood I will click on an article, tends to be 0.5% to maybe 1%, 2%.

Deeper Dive is being used by more than 10% constantly. So it's a double-digit adoption, which is -- I've never seen anything like that. And the CPM, like I mentioned, could be 5 to 10x. A single tiny ad is making such a great CPM, which is also why I believe Google is not only they're not concerned about blue link search. I'm convinced they just want search as we know it now to die as fast as possible and all move to Gemini because they probably know what I know now, which is ads on in conversations are worth a lot more than anything else. So that's really is now. And then our -- obviously, advertisers are excited to be part of LLM.

And about a month ago or 2, we got a lot of companies coming to us and say, "Can you help us monetize our AI because we don't have your ads, and it's very expensive." And we kind of launched Deeper Dive network. So if you have your own utility app and you're making it, you have a commerce AI app that you raised money for from investors and you want to make money, you can come to us and we'll show ads on your own LLM. So that's new.

And talk about the implementation. So how long does it take? Did it take or as an example, USA TODAY, to install it because essentially, what we're talking about is natural language search and conversational search across every single site that most every single company is trying to do. So talk to us about the implementation process. Maybe I'll back up. What gives Taboola the right to win here? Then talk about the implementation process and then where are we on adoption rate of your customers.

So the implementation itself is fairly easy. What we want to take a bit more time is we're always installing kind of like a QA process, a human and machine. We have AI as a judge agents that basically test the quality of those of your sites since we're happy with that. And then we have a human QA team that is just -- it's funny. It's like a whole new type of people that we've hired that are just news junkies. And they're just talking to, they just see how it feels like. So we tested out the total team is testing it out. We're launching on 5%, 10%, 50%, 100%.

So that part we want to take a bit more time because it's so new. We want people to feel comfortable with the process. So that is not complicated on the technical front, and there's an emotional part to it. And I'm not sure if you've ever driven a Tesla self-driving car. It's not about -- it's emotional for the first time, you're in the driving seat or your next to it, you can believe that you're not touching the wheel and maybe your kids in the back. So I think there's a self-driving car element here where publishers are saying, "Oh my god, this thing is answering on our behalf." So it's a bit of that experience for the first time, but then you love it, and they love it a lot more once they see how valuable that is to them. So that's fairly straightforward. And then...

Adoption rates.

Of publishers?

Yes, of your publishers that are using.

Yes. So it's now been adopted by a few dozens of publishers. And also, there's language barriers like we're launching it language by language, but it's a very fast-growing part of our business, still small in revenue, but we just -- we're about 10 million -- so about 10 million people are using it. So if you look at Claude, OpenAI, Perplexity, I think Perplexity is about 30 million people. So I want to be bigger than perplexity within a year or so in terms of adoption. And if we get more and more publishers using it, we just launched FOX News, which is amazing. We have such an incredible partner list. If we can get many of them to use Deeper Dive, my goal is to become the top 5 kind of enter engine in the world and also help advertisers monetize that experience, much like Gemini because OpenAI and Claude, I'm not sure it would be a successful in ad because I'm not -- I'm not sure he's going to sleep and he's dreaming about CPM. I don't think so.

But Gemini will be very good at it, and I think we'll be very good at it. So I kind of want to be the second after Gemini in terms of LLM monetization.

Well, there's a lot to go off here, but we have 11 -- 12 minutes left. So let me get to Realize which you mentioned earlier. Maybe let's explain to the audience like what is Realize. But then also, how is it accelerating or at least building the business here.

Right. We launched Realize about 1.5 years ago or earlier last year, which was essentially a relaunch of our advertising platform, which used to be a native ad platform. So native ads are a subset of the advertising market that we did a really good job growing that business from 0 to almost $2 billion. And then we wanted to go after a bigger dream of basically monetizing the entire Internet, not just a piece of it. We wanted on the advertising side, if you can give us display ads, vertical ads, motion ads, whatever you have, give it to us and give us -- and we want to support all types of bidding strategies much like a Google does.

And that meant to offer like a PMax completely autonomous opportunity on control and different strategies that are available to in a walled garden. And then we also imagine that 1 day publishers will say, "No, we don't want you to just monetize the bottom of the article, monetize everything, just take everything and monetize it." And just a month ago, we announced -- so that's Realize, Realize is our version of Google or Meta ad engine, we call it Realize. And the goal of it is to offer you everything you're doing in a walled garden, outside of the walls, it's like Taboola at the walled garden outside of the walls.

And all inventory. Or the inventory is available that...

But the vision of that was we want to have access to everything on the inventory front and wanted to support all types of strategies and formats that advertisers are used to whether that's agentic, which is fully autonomous or with more control. And we're 1 year in, 1.5 years into it, first of all, we're seeing accelerated growth rates for our business, which is great. We're generating a good amount of EBITDA, which is mostly converting to free cash flow, and we're buying a lot of shares. We believe in the story. We support the story.

And also, we're seeing evidence from the market, which is the most important that it's working. A month ago, NBC was a very large media company announced with us, it's my dinner today. That's salutary dinner. That's all of NBC news and TODAY.com to start will be monetized by Taboola, which if you told me a few years ago that I'm going to have an opportunity to work with such incredible partner and monetize everything. The only way to put an ads on that site, if you're not buying directly from NBC, which, of course, they can always do is through Taboola exclusive, everything.

So all first party, so is NBC doing any first-party sales?

They can sell their own whatever they want always. But that's it. Once they're done selling, they're going to -- there's no more AdTech stack. They're not -- there's no more, let me manage 10 SSPs, 10 DSPs. Those days are gone, one. One Google search, one Google, Facebook and hopefully, one company that can help simplify this complicated AdTech space for publishers and for advertisers. And I just can't -- that day is an amazing day. One, it's one partnership, but a really good one. And if we can do that -- and I did say publicly that this was 3x revenue roughly approx, which means that if we can get 30% of our company to do that, we've doubled Taboola. Yes. And so that is such a -- I'm so excited to become the backbone for the Internet as a vision to say, which is probably Realize, which is your question, give us whatever assets you have because you're going to get stuck on Google at some point. You're going to get stuck on Facebook at some point and come to us after.

And on the supply side, have access to everything exclusive beautiful first-party performing engine across the web. So we did the first one, and I hope to be able to share more of that.

Great. And then a Deeper Dive part of this partnership or that comes later potentially.

Later potentially.

Yes. And talk just a little bit about the supply side. So the advertisers that are on Taboola. Yes, I think you mentioned earlier about $100,000 spend scaled advertise was the comment. Talk about the supply side in terms of growth and just the demand, the demand of these advertisers for the inventory that you have.

So we have around 2,000 plus scaled advertisers and we were generally if we can we prefer to have more advertisers that we're always happy to see a big share of wallet from the same advertisers. But we like diversity. We like to see that number going up whenever we can. And I think advertisers are really find comfort with our investment and road map as it relates to Realize and Realize+, which is our, like I mentioned, like our PMax or agentic and the NBC news announcement was so well received by the advertising community because that's exactly what they want. They want our flexibility to be anywhere on the page that makes sense for them as a brand and makes sense for them from a performance standpoint.

On the supply side, we're seeing a really interesting diversity of growth into different areas. The Yahoo partnership, which is an incredible one. We've announced -- I think it's 3 years-ish now, give us access to different things like e-mail page, and other very U.S.-based, which we were very global. So that was -- that opened their eyes to our email providers. They need monetization and AI services, too. And then with Taboola News, we're seeing -- it's one of our fastest growth of Taboola revenue-wise and that's Taboola on OEM on devices such as Samsung, Xiaomi and others. And that now makes you feel about mobile, just in general mobile, mobile apps, mobile devices, just all this mobile space so that's interesting kind of source of growth for us. And the product, which we talked about as well as the full monetize everything. So we're seeing kind of growth in different directions. I would say, on the negative side, we're seeing AI impacting negatively some publishers.

So we're seeing search traffic going down. We're fortunate that most of our publishers are bigger publishers so they have a lot of direct traffic. But the search traffic is being impacted by that.

Impacted in that they're not getting the results from SEO?

They're not getting as much search traffic as they used to. So we're working with our publishers to help them navigate that like things like Deeper Dive, home page personalization, these are all initiatives that are there to increase engagement and traffic. But I think that's the -- there is some also negative that is going on and I think that's one opportunity for companies to innovate and double down on partnerships. That's how we're seeing it.

Maybe 1 last on the publisher partnerships. We talked about NBC. Obviously, you mentioned Yahoo!, that was a big deal a few years ago. FOX News, I think, is on there, too.

Just launched.

How do we leverage that to get more publishers onto the platform? Or is that something we should be looking for every quarter more publishers can join or we already have.

No. I mean there's never enough. There's never enough. I don't know that was a rhetorical question, and you know I would say that. No, we -- I think we'll -- when I look at the future of supply, I'm seeing 3 types of supply growth. One, upsells from within the same publishers like the NBC ones. I want to see 3x again and again and again, whenever I can. We've done such a good job spending a decade building trust with those -- they know we love them so much and they love us back. So that's one growth. The second one is what you call liberation. If you work with someone at Taboola, we hope to work with you one day. And we offer you so much more than money so we would consider us Deeper Dive, personalization, integration on Taboola news to get traffic just a variety of things and now full site monetization, if that's what you want.

So and FOX News is an amazing partner. I mean I can't tell you how happy I am because one they're great, but two, advertisers adore them, like it's such an amazing engaged audience. So the second one is more of that -- and the third one is, I always call it like what's the next Yahoo, not from that necessarily structured the deal more about, I think there are so many other great big-sized consumer companies that are thinking to themselves, how do we tap into performance advertising budgets and who is big enough to support us and who is not Google and I think that's a very short list. We're not the only ones, obviously.

But I think we're on that list, and I'm excited about talking to amazing companies that again, I never thought were in my TAM that are in our TAM.

So to that, actually, it's a good segue to, I believe, the team or you just hired a new Chief Business Officer. First one that the company has had, if I'm not mistaken.

So we've had SVP of Revenue. That's our first kind of like C-level person in the U.S. And that -- and so I know Krishan for a decade from his days with Linda Yaccarino at NBC. And the hardest job of a CEO and the hardest job of any company is building a culture that is a good one that supports the people executing and being happy and attracts other good people. So Krishan has such a vote of confidence coming from Amazon, coming from NBC and he's awesome, he's been with us for 6 months now, doing the work. So I'm very happy to have him on the team.

Yes. That's great. We have about 2 minutes left. I don't know if there's any questions in the audience. If not, think of a question and we can come back and keep going. Okay. We'll keep going here. Let's talk about the business. And so more on the profitability side, I think you have a margin framework that you're reaching towards in terms of growth. Just as the mix shifts away or, I guess, mix shift towards display, vertical video, Deeper Dive, just full page monetization. Talk to us about that long-term view ex-TAC, like how do you -- there's a lot going on here Adam that's pretty exciting.

Yes. In general -- so first, we optimize for growth first. So we do want this management. We want to have our ex-TAC, which is what's left for us after we shared revenue with our partners, which is the most important number for us because we want to do good deals. We wanted that to be in the double digits so we want to be first in double digit organically, consistently. So investors look at that and we can look at that and be happy with it. And then we always said that long term, we want to be higher double-digit growth rates. But that's kind of like our main priority.

From there, we have about 30% EBITDA margin, which we always -- Steve, our CFO, had always spoke about profitable growth. So we want to be a company that is not only able to grow faster take advantage of our unique assets, data, AI, people, partners, all those things. We also want to generate a healthy EBITDA that can support us continue to investing in the business and growing even faster. And in times like that, even buying a lot of the shares and owning more, which is just fantastic.

And we convert 60% to 70% of EBITDA. So I would say our AdTech margin tends to be between 35% to 40%, sometimes higher, sometimes lower, based on different things. But around that, and so I wouldn't be excited if it's 35%, I wouldn't be excited if it's 39%. It's in that range and that's what we like to do of that 30% converts to -- that's EBITDA conversion. And from that, 67% is free cash flow, which right now, most of that is being used to buy shares and this is great because we're able to develop Deeper Dive and Realize+ and full site monetization and still buy so many shares. So I'm very proud of the team for being hard working, executing and slowly progressing us towards where we need to be. We know where we want to be. We know where we are now.

Perfect. Well, Adam, I couldn't have ended in a better spot. That's a great way and overview of the business. Deeper Dive seeing greater adoption, more publishers joining the supply, Realize+ automating the process and scaled advertisers. So thank you for the time today. I appreciate the time.

Thanks for having me.

Great. Thank you.

Good day, and thank you for standing by. Welcome to Taboola's 2026 Second Quarter Earnings Conference Call.

[Operator Instructions]

Please be advised that today's conference is being recorded. I would now like to turn the conference over to your first speaker today, Aadam Anwar, Head of Investor Relations.

Thank you, and good morning, everyone. Welcome to Taboola's Second Quarter 2026 Earnings Conference Call. I'm here with Adam Singolda, Taboola's Founder and CEO; and Steve Walker, Taboola's CFO. The company issued earnings materials today before market opened, and they are available in the Investors section of Taboola's website. And I'll quickly cover the safe harbor. Certain statements today, including our expectations for future periods, are forward-looking statements. They are not facts and are subject to material risks and uncertainties described in our SEC filings. These statements are based on currently available information, and we undertake no duty to update them, except as required by law.

Today's discussion is also subject to forward-looking statement limitations in the earnings press release. Future events could differ materially and adversely from those anticipated. During this call, we will use terms defined in the earnings release and refer to non-GAAP financial measures. For definitions and reconciliations to GAAP, please refer to the non-GAAP tables in the earnings press release posted on our website. With that, I'll turn the call over to Adam.

Thanks, Aadam. Good morning, everyone, and thank you for joining us today. The second quarter was another important step forward for Taboola. We continue to execute and delivered results above our guidance across our key metrics despite dealing with 2 headwinds during the quarter. The first was a Google policy change that deprecated our explore more product and the second was our decision to remove low-quality publishers that were not delivering value for advertisers. Despite these 2 headwinds, I'm happy with our ability to beat our key metrics, accelerate growth and repurchase a lot of shares. More importantly, we had some large strategic wins that demonstrate meaningful progress against our long-term vision.

We expect these new wins to gradually begin contributing to our ex-TAC in the fourth quarter and to ramp more considerably in 2027. These tailwinds give us the confidence to raise our full year ex-TAC guidance to 9%. What gives me confidence isn't just the financials. It's the validation we got this quarter that our strategy is working. We're continuing to offer advertisers a viable option beyond search and social while investing in our tech to drive advertiser success and strengthening our relationship with some of the world's leading publishers. Together, these reinforce our confidence in our path forward to sustainable double-digit ex-TAC growth. Before getting into more detail, let me remind everyone who we are and how we compete.

Taboola is one of the largest performance advertising companies outside of search and social, referred to as the open web. Similar to how Google and Meta understand intent within their own platforms, Taboola understand intent across the billions of consumers who read, watch and engage within trusted OEMs, apps and publishers across the open web. We then convert these signals into profitable and measurable outcomes for advertisers. That proprietary intent data and the AI-driven conversion machine we've built, that is Taboola. Now in a world where AI is evolving so fast, I believe the winners will be those with either unique data that LLMs cannot get or access to unique supply and distribution. Taboola has both.

Turning now into 2 strategic milestones that further validates our realized strategy. First, we expect to announce a first-of-its-kind expansion with one of our largest existing publisher partners, a premier media and entertainment company. This marks an important evolution for us, expanding our role from monetizing individual bottom of article placements to monetizing everything, including display, vertical formats, native and more. To put this opportunity into perspective, we estimate that display advertising alone on this publisher represents 2 or 3x the revenue of the traditional native placements we've historically monetized.

This is important for 3 reasons. First, it's a validation of our realized products and strategy built with the purpose to expand wallet share within our publishers by moving beyond native ads to handle the full suite of ad placements needs. Second, we believe it will demonstrate how publishers can move away from relying on multiple Adtech providers and now consolidate it all into a single partner. By doing this, publishers can reduce complexity, lower operational burden, improve efficiency and drive stronger revenue outcomes. Lastly, this will create an opportunity for our advertisers to take advantage of even more premium supply, and we expect this to be a model for how things can be done with other publishers going forward.

This partnership demonstrates that publishers increasingly value partners that can combine AI, proprietary data and advertiser demand to drive better monetization. At the same time, we continue to see strong validation of our strategy through our ability to win some of the world's leading publishers. A great example is FOX News, one of the top 5 publishers in the U.S. We've already built a strong relationship with Fox Local, Fox Sports and Fox Weather, and the addition of FOX News represents a substantial growth opportunity and a significant expansion of our partnership across the FOX ecosystem. We believe this win reflects the investments we're making in Realize and our continued focus on helping premium publishers like Fox generate more value through performance advertising and AI.

We're encouraged by this highly competitive win and believe it will further validate our ability to continue taking share in the performance advertising market. Moving beyond our business wins, we've continued investing in our technology, particularly Realize, our performance advertising platform driving greater scale, better signals and stronger performance for advertisers. We believe the future of advertising will increasingly be powered by AI, moving from manual campaign management to intelligent systems that understand advertisers' goal, make decisions and continuously optimize performance.

That's the vision behind Realize Plus, our AI-powered optimization framework that brings to the open web, the kind of automation advertisers have come to expect from solutions like Google Performance Max and Meta's Advantage+. Since launching Realize Plus beta, more than 300 advertisers have already adopted the platform, and we're seeing encouraging early results as advertisers use AI to improve campaign efficiency and performance. We also believe AI will fundamentally change how advertisers interact with advertising platforms. particularly holding companies, agencies and large advertisers. That is why we built our MCP and cloud integration, which enable advertisers and agencies to plan, launch and optimize campaigns through natural language, conversations with AI.

While still early, we're encouraged by the momentum with a few millions of dollars of advertiser spend already flowing through the integration. We believe these investments position us well to lead the next generation of performance advertising and creates more value for advertisers across the open web. To wrap things up, we continue to execute across the business and raised our full year ex-TAC guidance. Importantly, we also delivered strategic wins that demonstrate progress against our long-term vision. We're also allocating capital with discipline. In the second quarter, we repurchased approximately 9 million shares for $41 million, continuing to return the majority of our free cash flow through buybacks. We've repurchased approximately 20% of our outstanding shares since the beginning of 2025, while maintaining the right balance between investing for growth and returning capital to shareholders.

As we look ahead, we're excited about the momentum we're building, the actions we've taken and the initiatives we're putting in place are positioning us well for the back half of the year and into 2027. We're building a stronger, more durable business and are excited about the path ahead as we continue building the leading performance advertising platform for the open web. And with that, I'll hand it over to Steve.

Thanks, Adam, and good morning, everyone. We're pleased with our performance in the second quarter. We continue to execute against our strategy and delivered results above our guidance across our key metrics. In the second quarter, revenues grew 2% year-over-year to $476.8 million. Revenue was below our guidance this quarter, primarily as a result of 2 factors. The first was our continuing effort to optimize supply quality. As part of our ongoing focus on improving the quality and performance of our publisher network, we took a more aggressive approach in the second quarter by exiting publisher relationships that did not meet our standards for advertiser success. Because this should improve advertiser success across our network, we believe this will improve long-term revenue despite the negative impact on 2026 revenues.

The second factor relates to the impact from Google's policy changes that affected our Explore More feature, as Adam described earlier. This feature enabled users to discover additional sponsored content from a publisher site after they clicked on the back button. However, due to Google's policy change, we were no longer able to provide that product starting this quarter. Despite these headwinds, I was happy to see that the number of scaled advertisers on our network grew 2% year-over-year, though we did see an impact from the headwinds on our average revenue per scaled advertiser, which remained relatively flat. Ex-TAC gross profit increased 12% year-over-year to $192.4 million in the second quarter.

Growth in ex-TAC gross profit outpaced the growth in revenues due to a combination of factors. First, given the reduction in supply due to our network cleanup and the deprecation of Explore More, we saw an increase in ad rates, which drives higher ex-TAC margins. Second, we had a shift in the mix of our business towards higher-margin areas, partially driven by those same cleanup efforts. Our strong ex-TAC growth also reflects the continued scaling of Realize, along with strong contributions from Taboola News. I would note that if it were not for the Google policy change that affected our Explore More product, we would have exceeded the high end of our ex-TAC gross profit guidance.

Gross profit for the quarter was $139.5 million, up 3% year-over-year. Growth in ex-TAC gross profit contributed to this growth. This growth was partially offset by a onetime noncash write-down of approximately $12 million related to certain publisher prepayments that we no longer expect to recoup, which obviously does not impact the long-term economics of our business. Net income for the quarter was $4.3 million, with non-GAAP net income coming in at $41.3 million. Adjusted EBITDA for the quarter was $55.5 million, which was above the high end of our guidance and represented a margin of 29%. This reflects our ongoing discipline in expense management while continuing to invest in strategic priorities to support our long-term growth. Foreign exchange continues to be a headwind in 2026.

On a constant currency basis, FX represented roughly a $7.5 million headwind to second quarter adjusted EBITDA. Excluding this impact, adjusted EBITDA would have been approximately $63 million, which would have represented an adjusted EBITDA margin of 33%. We expect FX to remain a headwind for the remainder of 2026. In terms of cash generation, we had $31.3 million in operating cash flow in the second quarter and free cash flow of $17.3 million. We continue to expect to sustainably convert free cash flow from adjusted EBITDA at a 60% to 70% rate over any typical 4-quarter period. Turning to the balance sheet. We remain in a strong financial position. We ended the first quarter with a net cash balance of $61.1 million. Cash and cash equivalents totaled $133.1 million, which more than offset our long-term debt of $72 million. As of June 30, we had approximately $198 million of available liquidity under our $270 million revolving credit facility.

In the second quarter, we repurchased approximately 9.4 million shares at an average price of $4.42 for a total consideration of $41.4 million. We have approximately $114 million remaining under our authorization and continue to view share repurchases as a compelling use of the majority of our free cash flow. Moving to guidance. For the third quarter, we expect revenues to be between $460 million and $473 million, gross profit to be between $148 million and $152 million, ex-TAC gross profit to be $184 million to $190 million. Adjusted EBITDA to range from $51.5 million to $56.5 million and non-GAAP net income to be $38 million to $42 million. Reflecting continued momentum across the business, we are increasing our full year outlook for ex-TAC gross profit and adjusted EBITDA while also updating our revenue, gross profit and non-GAAP net income guidance.

We now expect revenue of $1.93 billion to $1.96 billion and gross profit of $605 million to $615 million. Importantly, we are raising our ex-TAC gross profit guidance by $7 million at the midpoint to $772 million to $783 million and raising adjusted EBITDA guidance by $3 million at the midpoint to $228 million to $240 million. We expect non-GAAP net income to be between $168 million and $176 million. Our updated revenue guidance incorporates forward-looking effects of the revenue impacts from our publisher network cleanup and the deprecation of our Explore More product due to Google's policy changes. I would also note that while there has been significant public discussion about the reduction of display ad impressions at open web publishers, our guidance reflects the impacts of these user behavior changes.

Our raised ex-TAC gross profit guidance is notable, given that our outlook now incorporates the impact of the deprecation of Explore More, which was expected to contribute over $20 million of ex-TAC in the second half of 2026. In summary, we continue to make meaningful progress against our strategic priorities. This quarter, I was particularly excited about the strengthening of our publisher network. Adding FOX News demonstrates the continued strength and growth of our network of exclusive supply. Our soon-to-be announced expansion with one of our larger existing publishers to full page monetization is a significant validation of our realized strategy and our expansion beyond native advertising.

While we are in the early stages of many of these initiatives, we are encouraged by the momentum we're seeing and believe our disciplined execution reinforces our confidence in returning to sustainable double-digit growth. With that, let's move to Q&A. Operator, can you please open the line for questions?

[Operator Instructions]

Our first call comes from the line of Naved Khan of B. Riley Securities.

Maybe just on the deeper dive. I think last time around, you updated 7 million or so daily active users. Can you maybe just update us on how that user engagement looks like currently versus the last update that you had? And then on Realize Plus, it looks like a good number of advertisers have adopted it. Can you just maybe talk about the advertising budget allocation? And also, I think you mentioned superior ROI and efficiency in that ad spending that goes through Realize, but just can you maybe put some numbers around it and give us a sense of how that is?

Thanks for the question. I can start. So with regards to deeper dive, we're -- I would say a few things. First of all, we're about to cross the 10 million kind of users, which is really astonishing growth rate. We launched this product in September of last year. So this is really encouraging to see publishers adopting it, but even more so, consumers using it when they come to visit publishers in a growing pace. We're seeing north of 10% of people using deeper dive when they land on publisher sites. So if you go to ES today, 1 in 10 or more will type a question or replace a suggested question and then start engaging with an AI mode on a publisher site, which has trusted content.

What's even more interesting to me is that we see the reaction from the industry. When publishers are thinking about the future, publishers know that the future is not going to be driven by traditional page views. It's going to be driven by conversations and LLM monetization and a much deeper relationship with consumers that can grow the ARPU of their business. And in that future, Taboola plays a much more strategic role because it's more than just a widget on a page. It's more just visible CPMs and things, and it's more about revolution and AI engagement. So we're seeing publishers choosing Taboola. I mentioned FOX News, which is such an exciting competitive win for us. And there's so much more to talk into that I'm excited to share hopefully later in the year.

So deeper dive is a differentiated kind of position for us in the company. And then when you talk to advertisers, LLM monetization is almost like the next CTV for them. CTV is a more mature market. LLM is at the beginning of it, and it's growing really, really fast. So for agencies and big advertisers to be part of the conversation and monetize that is really critical. And with deeper dive, we're getting in the room with agencies and advertisers. And the performance we're seeing for deeper dive is out of control. I always joke that I'm sure that when search -- when Google launched search ads, I don't know, 20 years ago, they probably were shocked by the gap between a traditional ad to a search ad. And that's what I'm seeing at Taboola.

The gap between traditional kind of Taboola monetization, which is great to what we're doing on deeper dive is quite significant. About Realize Plus, I think we shared that we have about 300 advertisers using it, which is, again, good to see that more advertisers are playing with it. We believe, again, that's going to be a big part of our business in the future. If you compare that to PMax and Advantage+ for Meta and Google, we think advertisers want that product and the opportunity to improve ROI for them is more significant because we have full control over how it's been utilized. So too early to go much deeper than that, but I'm encouraged by the amount of advertisers using it. And I think once the Google and Meta, this will become a bigger portion of our business.

Our next call is from Martin Crockett of Rosenblatt.

I wanted to maybe explore more about Explore More. Could you tell us a little bit -- you said $20 million impact. Is that revenues or ex-TAC gross profit or EBITDA? That's just one on the financial. And then second, if you could just give us a sense of the degree to which you have other exposures to things that might be subject to Google kind of quality controls. I was thinking of you guys namely doing ads on publisher websites, not exposed to kind of traffic flows like this. So if you could elaborate on that, that would be helpful.

Yes. So I can start with the product impact and then Steve, feel free to jump in. So we do think this is a onetime event from a Google perspective. They made the decision to do it. They kind of executed it faster than we had anticipated. Usually, Google at times will announce something and take months, years to actually do it. This one was faster. I assume it's impacted everyone, not just Taboola, but as it relates to us, we had a product that on publisher side when consumers click back button, a certain experience would come up and show mostly content and some ads. Google deprecated that kind of experience, which impacted in our world, something we call Explore More. We did come up with a new product. Engage, which is basically aiming to capture a lot of that revenue back in other ways within what the policies of Google.

So that's been rolled out, and I expect it to create growth in the future. I don't know if it's going to bring back 100% of the Explore More, but I think it has a chance of bringing a lot of it back. But it was a onetime event. It's in the guide, and I don't expect that type of thing to happen again. But of course, it's Google.

And in terms of your question, that $20 million -- over $20 million in the second half, that was ex-TAC.

And so we would have that in the second half and then in the first half of next year as well?

Correct. Yes. I mean that is -- over $20 million was a second half effect. So it will affect us in the first half from a comparison basis.

Okay. And then you guys are also talking about cleaning up some of the secondary publishers. You didn't really size that. Is there any sense of the size, revenue and/or ex-TAC impact of that? Or is it just much less material and so not really discrete breakout potential?

Well, I guess what we said is that between that and the Explore More, that was -- that made up the majority of the shortfall that we had on revenue. So you get -- you can kind of get a sense of the impact with what we've given on the Explore More plus that. But I think, generally speaking, the way we think about that is that's a short-term hit, long-term gain because ultimately, if you have supply in your network that's not performing for advertisers, it hurts your overall advertiser performance and you probably lose budget. Some of them don't even know that they're losing them because of that, but you're losing budget. So while it reduces the short-term revenue, we think it's a positive thing for our network over time. And Therefore, it should lead to better results in the longer term. So it's a kind of short-term pain, long-term gain type of situation.

Okay. And outside of these kind of discrete actions, just to reiterate, are you -- what's your sense of kind of the broader kind of macro for kind of ad flow across your network as you look into the fourth quarter, how are you feeling about the environment?

So generally speaking, the environment has remained relatively stable. So similar to what we've been saying in past quarters, it's not the most robust advertising market you're ever going to see, but it's fairly stable. Like investors, advertisers are looking every day to see if we're at war today or if we're not at war tomorrow and what's going on with inflation. And so there's a lot of, I would say, skittishness out there. But so far, advertisers, especially our performance advertisers have continued to spend and continue to kind of operate their businesses as usual. So it's been fairly stable. But I think there's a lot of people just watching what's going on to make sure that they're not surprised by something.

Our next call is from Laura Martin of Needham & Company.

Yes. Just following up a little bit on Barton's topic. Why now? Why cut this now? And is there more to go in this low quality? That's my first one.

Yes. Laura, so I think the why now is we really do this on an ongoing basis. We're always looking at our network and trying to find parts of the network that are not working for advertisers and cutting it. Q2 was just a very unusual quarter in that we had a number of publishers grow really large very quickly that we had to cut because they just weren't working for advertisers. So it was an unusual quarter in terms of the volume of this. But the why now is we always try and do this as soon as we find pockets of nonperforming supply because you just don't want to be harming your advertisers that way. So we always do it. This quarter was unusual just in terms of the volume because of how fast some of those publishers grew with us.

Okay. Great. And then shares are weak right now, I think, in part because of the dependence on Google. So can you just walk through when Google makes a policy change like this 20 -- really $40 million hit over the next 4 quarters on Explore More. It sounds, Adam, from your answer earlier that maybe they tell you this is going to happen and then you guys have some time to adjust, but this one just they did much faster, so you couldn't adjust fast enough. Did I understand how the Google impact works in terms of timing?

Yes. So you have that correct. So basically, Google announced this in April, just before our previous earnings. And so we had heard about it. But usually, Google takes quarters upon quarters to actually implement these things because usually, they want comments from publishers and they want to make sure that they're not harming somebody or having secondary effects that they don't -- that they hadn't anticipated. So the example I'll give is third-party cookies, which Google announced, what was it, 3 or 4 years ago, they were going to eliminate and then they delayed and delayed and delayed and eventually said they weren't going to. So we heard about it in April, didn't expect it to happen that quickly. So we didn't actually adjust our guidance or anything as a result of it, but then we were surprised, as Adam said, by how fast Google moved on this.

So yes, you're correct about the kind of the timing and how that happened. Having said that, as Adam said earlier, I don't know of any other products we have that has that type of dependence on a Google policy. Obviously, our publishers have search traffic from Google, which is a Google dependency, but it's less than 5% of our U.S. page views. So there's less dependency there for us than most. So I'm not aware of any other kind of big exposures we have in that way.

Our next question comes from James Kopelman of TD Cowen.

The first one is for Adam. I want to ask about a deeper dive and the broader opportunity to capitalize on chatbot engagement. When you look at the broader trends with AI chatbots, how quickly are consumers adopting them or willing to adopt them directly on publisher sites? And what sort of time line are you contemplating in terms of this new type of engagement becoming a significant driver of both time spent and monetization on publisher sites. It certainly seems like a huge opportunity, but I'm curious how quickly large publishers are moving on this? And are there some advertising verticals where you think publishers will move most quickly or most slowly?

Thanks for the question. So let me start from the end. If I could transition half of Taboola to deeper dive traffic now, I would -- I mean as much of Taboola's traffic, if I could move to deeper dive, I would do it, which I suspect it's exactly what Google wants to do with Bloomx into Gemini. The CPM opportunity and the monetization opportunity today -- and we just got started. This is before innovative advertising units, and this is even more allowing advertisers to target that in more sophisticated ways. It's already now in the realm of like 5 to sometimes 10x. So every 1,000 impressions that Taboola serves today on publisher site versus every 1,000 impressions we get on deeper dive, it is uncomparable, and it's actually quite chunking.

So one -- for us and for the industry, I hope it moves as fast as possible. Advertisers wanted to move fast, publishers wanted to move fast, and we as the Bridge wanted to move fast. In terms of what we're seeing already, like I mentioned, about 10%, we're able to convert about 10% of the traffic into deeper dive once we launch it, which creates immediately almost around 10% revenue growth. So it depends on the publisher. So it's already accretive to the revenue the publisher can make, but it's still small. Our operation now is just how fast can we adopt it, how fast can we move it. And because it's new and the publishers are exploring, do they do it on their own? Do they work with us, editorial concerns they have, it's still fairly new. So some move faster than others. But I already see the impact of us offering that as an example, and seeing publishers choosing Taboola versus competitors.

It's because they know the future is no longer widgets. So they want someone that can help them enter the future, monetize the future and grow together. So I think it's already making a positive impact as it relates to competitive kind of wins that we're having. And you'll see more, I hope, later in the year that we're advanced with. And that also relates to advertisers who want to monetize it. So both fast, we were at 70 million users, I think, a few months ago, and now we're at 10 million. And we launched kind of like an ad network about a month ago. So it's a deeper dive kind of network for other LLMs because we're getting requests from many utility apps and other companies that offer LLM to their users to want us to monetize it for them.

You can imagine consumers are not going to have 50 subscriptions. So they want all these LLM services need ads and nobody wants to put a banner under LLM. So we're unique in our ability to provide advertising that is native and beautiful and relevant and make high CPMs. So I think this is -- this can be big, but we're trying to be always conservative with investors and try to just kind of set expectations and see how it goes.

And then I have a follow-up for Steve. I wanted to just go back to the AI topic. How are you thinking about the potential for Agentic AI to help drive efficiency gains within Taboola -- among Taboola employees? I'm curious if you have any color there on any internal beta initiatives that -- how they may be progressing or what you're learning? And then I have a second question on the 2Q factors. I'll -- I guess I'll just go ahead and ask that now. I was just curious, could you separate and quantify the impact of what I would see as 3 factors? Obviously, you mentioned the Google policy change and the dropping the underperforming publishers, but I think you've also mentioned potentially the impact of search referral declines.

I know those were in guidance. I'm just trying to tease out how much impact from each of those 3 factors. And specifically on the search referral declines from AI, I'm curious if that trend worsened during the quarter versus what you saw when you issued 2Q guidance 3 months ago.

Yes. So starting with the first question. So in terms of AI efficiency gains within Taboola, we have a lot of initiatives now that we're working on that. We have -- I forget what the exact percentage is, but a significantly high percentage of our code now is written or affected by AI. So obviously, we're getting gains in terms of productivity in our R&D and product management groups from AI. We also have initiatives throughout the rest of the organization to work on automating and streamlining processes using AI. We have people centrally who are working on that with our groups. And then we also have people individually within our teams helping to automate processes. So it's exciting. I mean I do see real opportunities here to have significant efficiency gains.

But I think it's a bit too early right now to talk about exactly where it gets to or to give you quantification on that. I'll also say that we're also trying to be cognizant of the fact that it's one thing to get efficiency gains from AI. It's a different thing if that only means that you're then paying Anthropic or somebody the same amount that you saved on your own people. So we're also trying to be smart there. So we're actually working on hosting our own models in-house and doing some things that will keep that cost mitigated because that -- I think companies that aren't thinking about that could be in for a bit of a shock in the future. So working hard at it. I see huge opportunity there, but a bit too early to start talking about specific numbers. In terms of the second part, you mentioned quantifying the different impacts.

I guess I'll kind of restate what we've said, and that's kind of all we're offering right now in terms of quantification of the different impacts. But what we said is if you look at the overall revenue impact or the reduction that we had in our guidance on revenue, that -- the majority of that was from the 2 factors, Explore More deprecation or the Google policy change that deprecated our Explore More product plus the cleanup of our network. The -- and then we also said that the Explore More was going to be over $20 million of ex-TAC in the second half. So you can also kind of do some back of the envelope math there to understand what the likely gross revenue on it was.

I will tell you that it's a fairly high-margin product. So it's not our 35% to 40% that the rest of our business is. It's a bit higher than that. So when you do your kind of quantification of that, you can assume it's a bit higher margin. Those were the 2 big impacts. So the third impact that you mentioned, which is the impact on search traffic to publishers from agentic AI and LLMs, that's a smaller impact for us because we have seen that -- or we've said in the past that less than 5% of our U.S. network, as an example, is from search. So it's a small impact. We are seeing an impact there. So I don't want to say it's nothing, but that's smaller than those other 2 factors.

Our next question comes from Tyler DiMatteo of BTIG.

I wanted to come back to the publisher point. Can we just talk a little bit about, I guess, the nature of those publishers that you were talking about in the headwind comments and I guess, the type? And then secondarily, I guess, how do you kind of think about the mix of publishers here by vertical, et cetera, as you look to shift to more premium publishers? I guess, obviously, things are changing. So I'm just curious how you think about that mix and type. And then my second question is, as I just kind of look at the geo breakdown of revenue, it seems like this is entirely an ex U.S. phenomenon in terms of like where the revenue is coming lower. Is that correct? And is there anything else going on there in terms of the geographic breakdown?

Tyler, Okay. So starting with the first question. So those publishers that we basically removed from our network that we deemed to not have good advertiser performance, those were international publishers. So they were -- I think they were -- a lot of them were in the Greater China region. And generally speaking, what those publishers are is they have low-performing traffic. Now that could be because they have bought traffic or other types of illegitimate traffic. It could also just be that the nature of their traffic is such that they don't have consumers who convert because, frankly, I don't care if it's fraud or if it's just a type of consumer that doesn't convert. If it doesn't work for our advertisers, we really don't want it on our network.

So those publishers were international publishers mostly -- largely in the Greater China region, and they were, again, low performing for our network, so we removed them. In terms of your question about kind of the mix of publishers that we're looking for, we've always biased towards premium publishers. We are always looking for kind of the biggest brand names. I mean that's one of the reasons we're so excited about FOX News because that is a great -- well, I mean, it depends upon where you are in the political spectrum, but it's a great brand in the U.S. It draws in consumers, and it is something that it's a brand you want on your network. So we always biased towards that. That doesn't mean we always end up or that we never end up with lesser brand names or publishers that don't perform, and that's why we're always looking to clean up our network.

In terms of the geo question that you had, like I did mention that the removed publishers were more international. But the impact that we're seeing kind of on revenue from Explore More, the Google policy change, that one is more global, but the network cleanup was more international, if that answers your question.

Our next question comes from Briana Diaz of Citizens.

Just going back on the lower quality advertisers, how should we think about the impact to revenue per active advertiser and the number of advertisers in regards to 2Q and maybe if that contributed to the slowdown in the growth from 1Q to 2Q and how we should be thinking about those 2 metrics going forward for the remainder of the year? And then just a second question, can you just elaborate on the strategic significance of just expanding from individual article placements to monetizing the full suite of inventory feels like that's a big step change. And what's the opportunity to expand that to other publishers online?

Yes. Good question. Thanks. So first of all, on the first question about scaled advertisers. So yes, Q2, the growth of especially the average revenue per scaled advertiser, but frankly, also the number of scaled advertisers was impacted by the network cleanup that we did as well as the Google policy change that deprecated our Explore More product. So both of those were impacted. Obviously, when you intentionally decide to put -- to reduce revenue on your network by cleaning up and removing poor quality publishers, poor performing publishers, that is going to impact your average revenue per advertiser. So it did have an impact. I was pretty happy to see, though, that our number of scaled advertisers still grew 2% year-over-year because that's -- as I've said in the past, having more scaled advertisers means we've got more relationships with advertisers that we can then continue to grow in the future.

So it's good to see that. But the -- both that number and the average revenue per scaled advertiser were impacted by those other 2 factors. Looking forward, what I want to see, and I've said this ever since we started releasing those metrics, I would like to see continued growth in the number. That is probably the best leading indicator for where we're going and how we're doing. So that one I want to see continuously growing. Average revenue per scaled advertiser, as long as it's stable around its current level and not declining a lot, I'm pretty happy with that as well because I've mentioned this in the past, as we add more scaled advertisers, they tend to drag down that average a bit because when they first scale up, they're usually at the small end and then hopefully, over time, we can grow them.

So it's okay with me if that stays relatively stable. I don't want to see declines in it, but I'd like to see it relatively stable as long as we're growing the number of scaled advertisers. That's what I'd like to see as we go forward.

I can take the second one. So the partnership that I hope to announce quite soon actually is one of our largest publishers and a name you know. And what's interesting to me is not only the growth -- the financial growth, which I mentioned, it's in the realm of about 3x bigger, it's more about -- so which basically means that we're sitting on this base of revenue that could be significantly higher, but just upselling up our existing relationship and trust with publishers for the last decade and do a lot more for them. But what's interesting here is that you're seeing the industry with -- there's so much going on, you're seeing publishers basically wanting to have less partners, deeper relationships, less cost, less complexity and more revenue.

And because we're already a significant portion of the revenue and we have a lot of direct demand and we have programmatic pubs that are connected, we're in such a unique position to just say, give us everything. In this case, it wasn't even our idea. They came to us. This is a relationship with a very senior person there who suggested that this might be a good idea for both of us, and we engage in that, model that and doing it. And interestingly enough, at the same time, we're now in conversation with other publishers. So I do think this could be an industry kind of change that publishers want to have less partners instead of having 5 to 10 Adtech SSPs and DSPs and wrappers and all these names have one that can just be a monetization layer for the Internet, which is really my vision for the company.

So much like Google owned search and Facebook owns social, if we can become the single most important partner for the open web and the monetization, the economics for the Internet, that's a big place for us to be. So I hope to continue to share those. It's financially meaningful, and we're starting with one of the best names we have as a company.

Your next question comes from the line of Mark Zkutowich with Benchmark.

This is Alex on for Mark. How much of your revenue that you're walking away from is minimum guarantee inventory that you've chosen not to renew? And what is your current revenue exposure to minimum guarantees?

Yes. Thanks for the question. So almost none of that revenue that we walked away from was minimum guarantee. We really use minimum guarantees mostly for premium brand name publishers that you would know, and that rarely ends up being bad traffic or poor performing traffic for our advertisers. So usually, when we do cleanup, it's not minimum guarantee. And in this case, that was true. It was almost no minimum guarantee traffic. And I think this past quarter, we said that about 13% of our TAC was paid out under minimum guarantees. That's where we are as of right now. And obviously, the trend is towards more rev share and less minimum guarantees.

Got it. And then just a question on contribution ex-TAC margin. As your revenue base indexes more towards more premium publishers, could you discuss the yield efficiencies you're capturing relative to the potential pressure you're seeing from shifting exposure towards these larger publishers?

Are you asking whether or not going towards premium publishers is going to impact ex-TAC margin in some way? Or I guess, how are you...

Yes, relative to perhaps some of the yield improvements that you're seeing?

Yes, understood. So first of all, I think what our belief is we've been in the kind of 35% to 40% ex-TAC margin range for a while. I think that is a good expectation for investors to have going forward is that we should be in that 35% to 40% range. Any given quarter, it could be a little bit higher, a little bit lower depending on seasonality effects that quarter, mix of business, et cetera, but I think that's a good expectation. I don't think that -- I think, first of all, we are probably gaining competitiveness as we win publisher deals. I think the FOX News signing is a pretty good indicator that we are winning more business from our competitors than we're losing.

And I think we're gaining competitiveness, which is a good sign, obviously, for future ex-TAC margins because that is what impacts how much we have to pay to get a publisher is how competitive we are. And I think over time, we expect to become more and more competitive. So we think we have an ability to get higher ex-TAC margins in general over time. So now having said that, I still say expect 35% to 40% margins because we also have some business where we're newer. So for instance, the partnership that Adam mentioned with a publisher where we're going to start doing -- monetizing all of their ad units, all of their display and vertical video and everything else versus just their native.

That is -- we don't know exactly what the margin on that is going to be. It's too early to really know where we're going to get to. But I don't expect it to be necessarily as high as our legacy business out of the gate. So we'll probably need some time to optimize that over time. But even having said that, it's -- as Adam said, the gross revenue potential is 2x to 5x what our native is. So even if it's a slightly lower margin, it still has an opportunity to more than double our ex-TAC on most of those publishers. So I would say, overall, the shift towards premium publishers is not the key factor in our long-term ex-TAC margins. It's really how competitive we are, and we think we're gaining there. And we think, therefore, we have an opportunity to do better over time on our ex-TAC margin.

This concludes the question-and-answer session. I'd now like to turn it back to Adam Singolda for closing remarks.

Thanks for being us, everyone, this morning. This was an important quarter for us. We raised our guidance, again, continue to validate our strategy through major strategic wins, and we're making meaningful progress against our long-term vision. We bought back approximately 20% of our shares since 2025, and we intend to continue returning the majority of our free cash flow through share repurchases. Thank you for your support, and we look forward to speaking with all of you and many of you in weeks ahead. Thank you.

Thank you. This does conclude the program. You may now disconnect.

Good day, and thank you for standing by. Welcome to the Taboola Q1 2026 Earnings Call. [Operator Instructions] Please be advised that today's conference is being recorded.

I would now like to hand the conference over to your first speaker today, Aadam Anwar, Head of Investor Relations. Please go ahead.

Thank you, and good morning, everyone, and welcome to Taboola's First Quarter 2026 Earnings Conference Call. I'm here with Adam Singolda, Taboola's Founder and CEO; and Steve Walker, Taboola's CFO. The company issued earnings materials today before the market, and they are available in the Investors section of Taboola's website.

Now I'll quickly cover the safe harbor. Certain statements today, including our expectations for future periods, are forward-looking statements. They are not facts and are subject to material risks and uncertainties described in our SEC filings. These statements are based on currently available information, and we undertake no duty to update them, except as required by law. Today's discussion is also subject to forward-looking statement limitations in the earnings press release. Future events could differ materially and adversely from those anticipated.

During this call, we will use terms defined in the earnings release and refer to non-GAAP financial measures. For definitions and reconciliations to GAAP, please refer to the non-GAAP tables in the earnings release posted on our website.

With that, I'll turn the call over to Adam.

Thanks, Aadam. Good morning, everyone, and thank you for joining us today. We're starting the year off strong with our first quarter results exceeding the high end of our guidance across all metrics. We're seeing continued acceleration in our growth, which gives us the confidence to raise our full year guidance across the board. We now expect excess gross profit growth of 8% while maintaining 30% adjusted EBITDA margins and strong free cash flow conversion.

As I said last year, we believe we've reached an inflection point with Realize driving advertiser success. I'm confident that this momentum gives us a clear path to double-digit growth over time. We're not there yet, but we're moving in the right direction, and I'm proud of the team executing against it.

In the first quarter, we repurchased approximately 7 million shares for a total of $23.5 million while continuing to invest in R&D to support our long-term growth ambitions. And including this quarter, we've now bought 19% of Taboola between 2025 and year-to-date 2026, which we're very pleased with. We plan to continue allocating the majority of our free cash flow towards share repurchases, which we view as our most compelling capital allocation opportunity.

Before getting into the details, let me remind you who we are and how we compete. Taboola is one of the largest performance advertising companies outside of search and social, referred to as the open web.

Similar to how Google and Meta understand intent within their own platform, Taboola understands intent across billions of consumers who read, watch and engage with trusted OEMs, apps and publishers across the open web. We then convert these signals into profitable and measurable outcomes for advertisers. That proprietary intent data and the AI-driven conversion machine we've built, that is Taboola. In a world where AI is evolving so quickly, I believe the winners will be those with either unique data that LLMs cannot get or access to unique supply and distribution. Taboola has both.

To execute on our mission in 2026, we're focused on 3 priorities: first, investing in our technology to advance Realize; second, with Krishan Bhatia joining as the Chief Business Officer, further verticalizing our sales organization around our ideal customer profiles, where we're seeing stronger retention and spend growth over time; and third, strengthening our brand.

As advertisers see stronger results on Realize, our ability to expand the budgets we manage continues to grow. In the first quarter, Realize drove increases in both scaled advertisers, those who spend more than $100,000 a year with us and the budgets we manage. Scaled advertisers grew 3.5% and average revenue per scaled advertiser grew 5%. As we scale, we benefit from more data, which powers our AI systems and drives continuous performance improvements, reinforcing our ability to grow budgets over time.

This progress comes from our investments we've been making across our technology, strengthening our user graph to better understand users across sites and devices, leveraging unique signals from Taboola News, high-intent content like product reviews, intent signals driven by a massive amount of people clicking on our ads, along with ongoing improvements to our bidding and core algorithms.

Just a few weeks ago, we introduced Realize+ our agentic framework for advertisers, something the team has been building towards for a long time. Meta has Advantage+. Google has Performance Max, and now we have Realize+. The idea is simple. Advertisers who want greater control, such as setting budgets by strategy, defining goals by geo and managing campaigns more hands-on can continue to use Realize. However, for those who prefer full automation, they can simply provide a budget and objective and Realize+ will take care of the rest, including audience targeting, creative generation, placements and continuous optimization.

What matters here isn't just simplification, it's performance at scale. By reducing operational complexity and improving outcomes, Realize+ reacts autonomously to the dynamic marketplace in real time, deciding and executing strategies which drive better performance outcomes. This allows advertisers to confidently shift more budgets into the system over time. That's how we grow. We want to make it really easy to use Realize and succeed.

Our second priority is our go-to-market, where we're building a more repeatable engine to grow our share of advertisers' budget. The foundation of this strategy is verticalizing, organizing our sales team by industry and focusing on clearly defined ideal customer profile, what we call ICPs. For Taboola, these ICPs are performance-oriented advertisers who prioritize measurable outcomes, require scalable customer acquisition and operate in mid- to low-funnel categories such as travel, health care, auto, personal finance and more. By aligning our verticalized teams to these ICPs, we develop deeper expertise, execute faster and stay focused on delivering advertisers outcome.

Lastly, on brand and perception, we're making real progress in how the market sees Taboola. As we invest in products like Realize+, onboarding incredible advertisers and partners, we're shaping our brand to be recognized as an AI-driven performance platform. This takes time, but we're building trust and shifting perception.

In the end, I measure this by outcomes. Are we breaking more advertisers, driving more demand and growing faster or not. At the end of the day, companies either accelerate their growth or they don't. And to bring it all together, we feel good about where we are and even more importantly, about where we're going. We're seeing early signs of what this business can become when technology, data and execution come together. It's still early, but we're moving in the right direction. It's an exciting time for us at Taboola, and we look forward to updating you all on our progress throughout the year.

With that, I'll hand it over to Steve.

Thanks, Adam, and good morning, everyone. We're happy to start the year on a strong note. In the first quarter, we continue to build on the momentum we built last year, delivering results that exceeded the high end of our guidance across every metric. In the first quarter, revenues grew 9% year-over-year to $466.4 million.

We remain focused on increasing advertiser investments through Realize, our performance advertising platform. Continued product enhancements and new feature launches contributed to solid execution during the quarter. This was evident in our first quarter scaled advertiser metrics, which showed a 3.5% rise in the number of scaled advertisers and a 5% increase in average revenue per scaled advertiser.

Ex-TAC gross profit increased 11% year-on-year to $168.1 million in the first quarter. Growth was primarily driven by higher advertising spend, largely supported by the scaling of Realize as well as strong performance from Taboola News and Bidded Supply. Gross profit for the quarter was $129.6 million, up 9% year-over-year. Growth in ex-TAC gross profit contributed to this performance, but was partially offset by an increase in infrastructure and operational costs as we continue to scale the business for future growth.

Net income for the quarter was $59.1 million with non-GAAP net income coming in at $17.2 million. Net income came in higher due to proceeds from a onetime legal settlement. This settlement was adjusted out of non-GAAP net income. Adjusted EBITDA for the quarter was $26.7 million, a margin of 16%. This reflects continued discipline in expense management while continuing to invest in strategic priorities to support long-term growth. And I would note that the legal settlement I mentioned previously does not contribute to adjusted EBITDA.

Foreign exchange was a meaningful headwind in the quarter. On a constant currency basis, first quarter ex-TAC gross profit showed a tailwind of approximately $3.6 million, while operating expenses saw a headwind of approximately $8.2 million, primarily reflecting the strength of the Israeli shekel, where we have a significant employee and cost base. In aggregate, FX represented roughly a $4.7 million headwind to the first quarter adjusted EBITDA. Excluding this impact, adjusted EBITDA would have been $31.4 million, which would have represented an adjusted EBITDA margin of 19.1%. We expect FX to remain a headwind for the remainder of 2026.

In terms of cash generation, we had $108.7 million in operating cash flow in the first quarter and free cash flow of $90.3 million. Free cash flow for the quarter benefited from the legal settlement I mentioned previously. As a reminder, we expect to sustainably convert free cash flow from adjusted EBITDA at a 60% to 70% rate over any typical 4-quarter period.

Turning to the balance sheet. We remain in a strong financial position. We ended the first quarter with a net cash balance of $83.9 million. Cash and cash equivalents totaled $150.3 million, which more than offset our long-term debt of $66.4 million. Last year, we secured a $270 million revolving credit facility. And as of March 31, we maintained approximately $203.6 million of available liquidity. We remain focused on disciplined capital allocation, prioritizing investments in sales and R&D while returning excess capital to shareholders through share repurchases.

In the first quarter, we repurchased approximately 7 million shares at an average price of $3.41 for a total consideration of $23.5 million. As a result, shares outstanding declined to approximately 273 million at quarter end, down from about 276 million at the end of 2025. We have approximately $160 million remaining under our authorization and continue to view share repurchases as a compelling use of the majority of our free cash flow.

Moving to guidance. For the second quarter, we expect revenues to be between $492 million and $505 million, gross profit to be between $147 million and $152 million, ex-TAC gross profit to be $189 million to $194 million, adjusted EBITDA to range from $49 million to $55 million and non-GAAP net income to be $36 million to $43 million.

Reflecting continued adoption of Realize and its features, we are raising our full year guidance across all metrics. We now expect revenues to be between $2 billion and $2.06 billion, gross profit to be between $610 million and $630 million, ex-TAC gross profit to be $760 million to $781 million, adjusted EBITDA to be $222 million to $240 million and non-GAAP net income to be $167 million to $191 million.

I would note that our adjusted EBITDA guidance reflects a forecasted headwind from foreign exchange rates of approximately $13 million in operating expenses, partially offset by ex-TAC tailwinds. Without this headwind from foreign exchange, adjusted EBITDA margins would be approximately 34%.

In summary, the first quarter results exceeded the high end of our guidance range across all metrics, reflecting strong execution and continued momentum in the business. We continue to build on the momentum we've seen with Realize and are focused on accelerating growth. We continue to stay disciplined in our approach and our steady progress reinforces our confidence in our ability to return to sustainable double-digit growth over time.

With that, let's move to Q&A. Operator, can you please open the line for questions?

[Operator Instructions] Our first question comes from the line of Daniel Medina of Needham & Company.

Can you hear me? It's Laura Martin, can you guys hear me?

Laura, we hear you, if you can hear us.

Okay. Yes, I just didn't know if you can hear me. Yes. So I have 2. One is on the scaled advertiser number, these numbers look great. Can you remind us like why you make this distinction between scaled advertisers and like -- and what is the -- is the churn level different with non-scaled advertisers? Or why do we make this distinction in scaled advertisers?

And then the other thing is, I think I saw it when we were talking last week, Adam, we were talking about your integration into Claude and how you sort of think that the -- maybe these LLMs are kind of a new source of demand for Taboola. Could you go into how you're thinking about some of these Agentic AI LLMs and whether you think that drives revenue growth for you in the future?

It was good seeing you last week at POSSIBLE in Miami. So with the first question as it relates to scaled advertisers. The reason we think that matters is, as a performance advertising platform, people try Taboola and some of them obviously succeed and some of them need more work to succeed. But what happens when someone kind of exceeds the $100,000 mark, they tend to be very stable line of revenue for us.

It means they've tested enough, they tried enough of our capabilities to feel good about the performance that they were hoping to get. And at that point, for us, that kind of becomes more sustainable, predictable line of revenue, and we can grow that base over time. So I think for investors, we think that's an important metric because it's a good proxy for, one, how are we doing as a technology platform? Are we able to grow that number? Are we able to get more and more clients to be happy with what we're seeing as that compares to Meta and Google. And two, that revenue is fairly predictable as it relates to churn rates and things like that, like you mentioned. So we think that's a good metric to track.

Internally, there are leading indicators that get advertisers to that stage. So usually, we lower churn rates, they're able to spend more money with us until they hit that point of scaled advertisers. So that's why we track it internally. And I think for investors, that's a good proxy for our progress as a technology company as well as how sustainable is that revenue moving forward.

About Agentic AI, which we talked a lot about last week, and I'm personally very excited about it. I think as an industry, we're going through a significant revolution with AI, not only affecting almost everything we see and touch, but now specifically with programmatic protocols. We're spending a lot of time with agencies and big advertisers. And for the last 30 years, they've spent tens of billions of dollars buying programmatic different types of supply.

And the challenge with programmatic protocols is that they normalize for the kind of the lowest denominator. They can't really take advantage of the unique data different companies have. They can't take advantage of the unique supply companies have. And with agents now kind of agent-to-agent era we're embarking, you can now -- with Taboola, you can go to Claude and using an MCP, basically a skill that is able to talk within the app or within the CLI, the command line, if that's what you're using, you can talk to Taboola Realize or you can talk to Taboola Realize+, never leave Claude and basically interacting your objectives.

The reason this is exciting is you can do the same with Google, you can do the same with Meta, you can do the same with Taboola and even TV, which means for a $200 subscription with Claude, you can now buy search, social, open web and TV. And that is quite big. So it's early days, but I think things will move very fast. And I suspect a year from now, Laura, as you and I do a fireside chat and talk about where things stand, I suspect agent-to-agent kind of advertising buying will be a much bigger portion of the industry.

Our next question comes from the line of Barton Crockett of Rosenblatt.

I was curious, you credited Realize was driving some of the upside in the quarter and in the guidance for the year. And I was just wondering if you could be a little bit more specific about what in Realize was driving the upside. Was that using Realize to perhaps go beyond some of the traditional bottom of page inventory that you've been trafficking in going into the other parts of the page? Or was it just more general kind of the capabilities that Realize that was driving? That would be one question.

And the other question is on the guide. You guys are raising the ex-TAC gross profit guide and revenue guide at both the high and low end, but the EBITDA net income guides are less changed. I mean, unchanged on the EBITDA at the low end and unchanged on the non-GAAP net income at the high end. And so why isn't the revenue upside flowing to the bottom line as much?

I can start, and thank you for the question. So on the first one, with Realize, essentially, we're really seeing utilization of all the various kind of capabilities the platform offers for advertisers being used more, which accelerates advertiser success, which accelerates essentially spend on our platform, and that's why we're able to raise our guidance and feel good about our way towards double-digit growth consistently and organically as a company. And that includes things such as format diversification, which you mentioned. So it's much, much easier now to start a campaign with either vertical video or display.

On the supply side, we're plugged into much more, I would say, kind of traditional display inventory if that's what advertisers want; vertical format, if that's what advertisers want. On our OEM, we have full screen kind of advertising placements, in-app inventory, which is about -- again over $100 million a year as well. So we're seeing basically on the supply also further diversification of types of inventory advertisers can get. And then one of our probably leading tech features that advertisers really use is predictive audiences, again. So advertisers keep using our ability to predict how many more conversions they can get based on the seed of conversion they already have with us.

And to oversimplify this, what that means if you're a personal finance mortgage company and you were able to get 1,000 leads with Taboola, we're able to predict how much money do you need to give us to get the next 1,000 conversion, which is really comforting for advertisers who are looking for stability and predictability with us. They want to know how much more money do they need to give us so they can further scale their spend and work with us on the platform. And that's something that advertisers really like.

And as you've probably seen with Realize+, this will be, I hope, and that's what I expect over time to see even further kind of acceleration of how advertisers use Taboola. With Realize today, advertisers open sometimes dozens or hundreds of campaigns with Taboola. And they have to manage that manually, which some of them really like to do that to have that level of control. But with Realize+, Realize+ may open for advertisers dozens, hundreds and sometimes thousands of campaigns for you on a daily basis, geo campaigns, different bidding strategies campaign, retargeting, different things.

So I hope to see even further automatic utilization of what Realize can do with Realize+. But all of those things as a mix are able to increase spend, grow the scaled advertisers and, of course, help us accelerate the guidance for the year.

Regarding -- hey Barton, regarding your second question about kind of the flow-through on some of our guidance, I think the biggest factor on why we didn't flow through as much of the beat on adjusted EBITDA and non-GAAP net income as we did on ex-TAC and revenue has to do with foreign exchange rates, primarily the Israeli shekel, in fact. So the Israeli shekel will impact our OpEx by about $13 million this year. It's a $13 million headwind. And that obviously has a big impact on that adjusted EBITDA.

We're happy though that even with that headwind, we're still guiding to 30% adjusted EBITDA margins for the year. Without that headwind, our adjusted EBITDA margins would be around 34%. So generally, it's related to exchange rates. Same thing on non-GAAP net income, except it's actually even a bit more extreme because we tend to hedge our cash expenses but we are not 100% hedged, unfortunately, because you can never be perfectly hedged. But we tend to hedge our cash expenses. We don't hedge our noncash expenses at all.

So therefore, higher exchange rates have an outsized impact on non-GAAP net income. So we're -- we always try and be a bit more conservative there and the flow-through is less because of that.

Our next question comes from the line of Tyler DiMatteo of BTIG.

Steve, 2 for you. My first one, on the guidance, how much conservatism is baked into that from a macro perspective, given everything that's going on in the world today? And then maybe how much of a contribution from something like live events? I know in the past, I believe you had said live events is maybe more of a traffic boost than an actual revenue boost per se, but just kind of curious on those 2 things.

And then the second question for you, Steve. Has the time line to double-digit growth? Or I guess, how has the time line to double-digit growth changed at this point?

Yes. So good questions. Thanks for asking them. So I think, generally speaking, when it comes to the macro, it's been actually pretty impressive that the advertising marketplace, especially our advertising marketplace in the performance space has been fairly resilient in the face of wars, tariffs, everything that's going on kind of in the macro, it's continued to be relatively resilient. So I find that pretty impressive, and we continue to say that it's been a fairly stable marketplace. So that's good.

When it comes to the events like World Cup and the elections and things like that, I think we've mentioned this in the past, and you're right in your characterization of it, which is that it's more of a traffic event for us than it is an advertiser interest event for us. The reason for that is that those events tend to be more branding oriented. So in elections, it's usually the candidates trying to get their names out there and send branding messages, either branding themselves or anti-branding their opponents.

We do get some flow-through from that. Usually, it's more around like campaign contributions and very specific performance actions that they also want to achieve. But generally speaking, it's much smaller impact on us than it is on, say, connected TV marketplace or something along those lines. Same thing for World Cup. It's usually companies like Coca-Cola trying to build their brand by tying themselves to those events. So it usually is more of a traffic impact for us with a small amount of flow-through on the advertising side.

Then to your last question about time line to double-digit growth, I don't think -- we feel good. I think Adam mentioned in his prepared remarks that we still feel good that we're on the path to consistent double-digit growth. We still feel like we've seen an inflection in our business, and we're continuing to make progress towards that. Obviously, you can see in our guide that we're not there yet. So Adam is not happy yet about where we're at, but we're making progress. I wouldn't say anything has changed on the time line with anything we've seen recently.

Our next call comes from the line of Brianna Diaz of Citizens.

This is Brianna on for Matt Condon. Can you just unpack the outperformance in the quarter? And what were the contributions of growth from new products such as Realize and DeeperDive? And if any at all growth is embedded in the full year guide from those products?

And then on Realize+, it's just doing more of the live work for an advertiser. Is there anything to know on just the take rate and pricing and how that might compare to a traditional campaign?

I can start, Steve, feel free to join. So in terms of the contribution right now, it's primarily driven by our strategy to make advertisers successful and spend more money with us and get more advertisers to work with us. So most of what you're seeing now in the business is directly correlated to us making more advertisers successful and existing advertisers spend more with us, which is exactly kind of tracked through the scaled advertisers you're seeing, 3.5% more -- numerically more advertisers, and then you're seeing increase in average spend per advertisers. So that's primarily Realize, which is most of our revenue as a company.

Again, one of the things that's unique about Taboola, when you look at our $2 billion of spend, gross revenue, the vast majority of it is direct to Realize. So it's not programmatic. It's not through channels, it's advertisers buying from us, much like they are buying from Meta and Google. And to me, that's a very unique part of our business and the company.

DeeperDive, which you mentioned, is growing really fast. I mean it's quite -- we just had a Board meeting yesterday, and we talked about how much fun it is to kind of start a start-up within a start-up like DeeperDive completely organically, having a small team of really kind of ventures working so hard to bring kind of a ChatGPT-like product for the open web. And I would argue in doing things that they can't do like suggesting questions based on first-party data that we have, and it's been used by incredible partners like you said today, Nexstar and Huffington Post and BuzzFeed and Independent and Reach and many -- and that business is growing.

Financially, that's still small, though I did mention that what we're seeing is unique. When I look at effective CPMs on a DeeperDive page or when I look at advertiser conversion rates, the return on ad spend from DeeperDive compared to anything else, it's at the top. So if DeeperDive continues to scale, I'm excited about the impact it can make to our publishers, the impact it can make to advertisers and the effect it can make to us. You may also know that I'm fairly optimistic about Gemini and Google given what we're seeing in DeeperDive, I suspect Google is seeing similar kind of trends for Gemini. So that's about that.

About Realize+, it's -- yes, you're right. It's basically making it dramatically easier for advertisers. By the way, also big advertisers who may have a Realize kind of operation, but also want on top of that to have a PMax-type line of business with us. So Realize+ can work very well for huge advertisers, big advertisers and smaller advertisers. But the idea is that the amount of permutations of Realize that our product team has brought to market is quite significant. There's a lot of different things you can do, more than a team of 1 or 2 or 3 humanly possible can execute on.

But Realize+ is unlimited. It can do everything for you. It doesn't sleep 7 days a week. So we're optimistic about where it can go. But of course, early days, and we'll continue to update.

Our next question comes from the line of James Kopelman of TD Cowen.

The first one is for Adam. Taboola has been a company that's benefited over time from acquisitions and some really large-scale partnerships. Obviously, Connexity a few years ago, and then the Yahoo! agreement, Apple News and other large partnerships. My question is, do you see Taboola's growth story as largely organic going forward? Or do you see potential opportunity for additional acquisitions over time or partnerships in new verticals?

And what sort of adjacent or additional competencies would Taboola potentially look to add over time as you continue to grow and look to capture share -- greater share of digital advertising? And then I'll have a follow-up for Steve.

Yes. That's a great question. So I'll say 2-part answer. One, I think most of our growth will continue to be organic. And at the same time, I'm quite happy to buy 19% of the company back since last year. So I like that we're not only accelerating growth, not only I see organic -- I keep saying consistent and organic. I use those words because that's how I expect us to continue to grow organically and consistently, which is something we care about and I think investors should care about. So that's how we see our future.

So when I say double-digit growth, I mean organically and consistently. It doesn't mean we're not looking at stuff all the time, but our appetite for a big kind of type of thing is small. And at the same time, like I said, I enjoy reducing the share count as a shareholder myself. I like that we're able to use our free cash flow, which is growing to put that to work, and we think that's a great investment for us. So that's something we intend to continue to do this year.

Now in terms of -- I can tell you, we had yesterday, our COO, our Chief Business Officer, Krishan and I spent some time together. And we're thinking about big growth engines over the next few years. It's kind of divided into 3. One is on the business side, there's a lot of sales growth with agencies and advertisers and partnerships. I think there's a lot of companies who want a really good friend, who want to drive -- build an advertising business for them.

We're obviously so happy with what we've done with Yahoo! and Apple and Microsoft over the years, but I think there's going to be more of those, and we're already in conversation now with some really exciting companies, big companies that want to -- they want a non-Google, Facebook friend, and we are the best friend they can have, and they can be the best ones for us.

The second thing is on the technology side. You're seeing us kind of investing a lot with Realize and Realize+ and more things coming up on the road map because we think that Taboola as a technology company can do a lot for ourselves by making advertising more successful and growing the ARPU or kind of like revenue per publisher or partner over time. So that's -- and our pipeline on the publisher side has never been stronger. I mean the meetings we're having, the seniority of people that take meetings is so great to see. And I think it's driven by DeeperDive and the strategic things we're doing that they want to see what we're working on.

And the third one is just AI in general. Our COO is leading an internal kind of huge multiyear project starting now, whereby we're constantly imagining Taboola as what we refer to as AI native company. What would we look like if we started today and constantly looking for innovation and growth and how we can all be more productive internally and externally. So between those 3 kind of big waves of growth, organically and consistently, I'm quite excited.

Great. And then a quick follow-up for Steve. I think Taboola has currently a little under 1,600 employees, if I understood that right from the 10-Q. That's down quite a bit from, I think, about 2,000 employees roughly a year ago. So I assume you're seeing some efficiency gains.

And I'm wondering how much of the efficiency is tied to AI initiatives or other areas within the company? And I guess, looking ahead a year or 2, what can you tell us about potential headcount trends, maybe areas of hiring as you continue to ramp Realize and how you balance those investment opportunities around Realize against the focus to remain prudent with regards to cost discipline going forward?

Thanks, James. Yes. So first of all, just on the numbers side of things, we're actually more like around 1,950 employees, right now. So I'm not sure maybe that was a subset of our company or something that you were looking at, but we're around 1,950. That's still down from where we were. So we were over 2,000.

We did just do a kind of an adjustment to our restructuring to our company that impacted that. Frankly, that was a relatively business as usual type -- ordinary course of business type of adjustment. So we were just reducing some investment in certain areas that we didn't want to invest in as much anymore. We're still hiring in other areas. That one was not as much about AI, although we are always looking at, as Adam said, areas where we can use AI to become more efficient.

And I don't want to necessarily predict the future too much because our future headcount growth and everything else is dependent partially on how we grow as a company. But what I would say as I look forward is, for sure, if I'm asking myself, can we be more efficient with the same number of people? My answer is yes. I mean, ultimately, our R&D group, for instance, is talking about a 10x project where they use AI to 10x the impact of their engineers.

It doesn't mean we necessarily see needing less people, but it means we can become way more efficient with the people we have, and we can support growth at a more efficient level. So I think the trend line without getting into specific headcount numbers, I think, is towards more efficiency. And I really -- we're all very excited about AI and the impact it can have on kind of our cost efficiency and everything going forward.

Our next question comes from the line of Naved Khan of B. Riley Securities.

A couple of questions from me. So maybe just on Realize, can you just talk about the ads -- the spend per advertiser on Realize and how it compares to the legacy native? And are you adding more verticals besides the one you've mentioned in the past, like finance, travel, et cetera? So that's one.

And then maybe just talk about the costs related to AI. So as you put in more AI features in your products, how should we be thinking about the impact from a cost perspective as the adoption increases for these products?

Sure. Let me -- I'll jump in and answer this. So when it first -- on your first question about Realize, I think what we're doing, we've talked about in the past that we're focused on 3 things, which we think will impact our growth with Realize.

One is, we're focusing on ICPs, ideal customer profile verticals. I wouldn't say we're adding more of those at this point. For right now, we're basically trying to focus our organization on making the ones that we've identified successful. So that includes tech efforts to try and make them work. It includes focusing our sales teams to make sure they're going after the right verticals. And the result of that, we believe, will be higher retention and more spend from our advertisers in general, especially within those verticals.

The second thing we're doing is we're investing in our brand. So we're trying to reposition ourselves from being a native company to being all performance advertising. So that is an ongoing effort. I think we're early in that process, and that's something we're continuing.

And then lastly, we're investing in tech, which I think Adam addressed a bit earlier. A lot of good things going on there. We're continuing to see opportunities to build on our product and to continue to develop more capabilities for our advertisers. Realize+ being a great example of that where we're making it so much easier to buy with us. We also released a skill in Claude now where you can interact with Realize directly in an agent-to-agent way. So things like that are going to drive growth in the future for us, and that's where we're focused.

To your second question, if I understand what you're asking, I think, generally speaking, AI is an opportunity for us to do more with less. So it doesn't -- and again, I think what that means is that we should become more cost efficient going forward. That includes the cost of the AI itself. Obviously, we have to pay for the AI, but we're also trying to be very smart about that.

So for instance, in many cases, we're hosting our own AI. So we bring in the open source AI models. We host it on our own infrastructure, which fortunately, we've built the company from the beginning to host our own infrastructure. So we host the AI models on our own infrastructure. What that means is you may not be up with the absolute latest model from whichever model you're using, but it's a lot less expensive to host it yourself. So generally, AI is going to drive cost efficiencies even after the cost of the AI itself.

This concludes the question-and-answer session. I would now like to turn it back to Adam Singolda, CEO, for closing remarks.

Thanks, everyone, for being with us this morning. Q1 wasn't just about beating the numbers. It's another step towards building the largest kind of walled garden outside of the walls, helping advertisers drive outcomes on the open web through Realize and now through Realize+, while growing our partners across publishers, apps and OEMs.

We're executing on our priorities, raising the guidance with confidence in our path to double-digit growth organically and consistently. I love that we're able to buy 90% of our shares since last year, and we do intend to aggressively keep buying shares this year.

We appreciate your support and looking forward to staying in touch these weeks ahead. Thanks, everyone.

Thank you for your participation in today's conference. This does conclude the program. You may now disconnect.

All right. Before I get into it, I'd just like to read an important disclosure. Please see Morgan Stanley research disclosure website at www.morganstanley.com/researchdisclosures. If you have any questions, please reach out to your Morgan Stanley sales representative.

Good morning, everyone. My name is Brandon Feldt. I lead up our digital advertising practice in the investment bank at Morgan Stanley. I'm joined here today by Stephen Walker, CFO of Taboola, which is a leading independent ad tech platform. Welcome, Stephen.

Thank you. Thanks for having us.

Of course. So to kick things off, for anyone who may not be familiar with the Taboola story, how would you describe the company today and how that description has evolved over the last few years?

Sure. So we're the leading performance advertising platform for the open web. What that means is we help deliver search and social-like performance outcomes for advertisers, but in the open web. So billions of consumers read and watch and interact with content on trusted publisher websites out in the open web every day. Our job is to find what we call moments of intent within that activity where we can help deliver performance outcomes for advertisers.

So Google knows intent on their platform, Meta knows intent on their apps and services. We know intent on the open web. And the way we know that is we have significant scale. So we reach 600 million consumers every month. We have very unique intent data because we see what people are reading, what they're clicking on. We even have pixels downstream to see what they're buying and converting on. And then we also have technology, AI-driven technology that helps match that very unique data with the intent signals to put the right ads in front of the right consumers at the right time.

So bottom line, we help advertisers deliver basically reach audiences in the open web with good performance outcomes. And in so doing, we also help publishers then monetize their unique audiences on the open web. So that's who Taboola is.

Awesome. Before we jump ahead to where Taboola is going, can you just take a step back and recap 2025 for us? What were the most important takeaways investors should understand to frame the conversation today?

Yes. So 2025 was a good year for us, very -- is a transitional year in many ways, but we ended up delivering almost $2 billion of revenue, $714 million of ex-TAC, which is the revenue we keep after we pay our publishers. $214 million of EBITDA, about a 30% EBITDA margin. So overall, a very strong year financially. And I think the big story for 2025 was the rollout of our new advertising platform called Realize. And what Realize did is it took us from being a native advertising company.

Those are kind of those ads that look like publisher content that you'll see on publisher sites to being really performance anywhere. So we can do display, we can do vertical video. We can do all formats, all placements and really help advertisers across a much broader area. We rolled that out early in 2025, and that was really what we focused our attention on. And we saw good progress over the course of the year in rolling out Realize. The other thing I would note is we're generating very strong cash flow right now. So we had over $160 million of cash flow last year. We bought back about 18% of our shares last year.

Terrific. Anything about 2025 that surprised you? There was a lot of volatility in markets, broader economic backdrop. It sounds like the business still performed very strongly.

Yes. I wouldn't say there was anything that was particularly surprising. I think we were pretty happy with the reception we got to Realize. So we were making a pretty major transition, right? So we're going from being fairly specialized in the online advertising world in terms of native. We estimated that, that was probably a $4 billion to $5 billion market to really trying to be performance anywhere on the open web. And we think that's 10x that size market. So that's a $40 billion to $50 billion market instead.

So that was a -- it's a big step for us. We consider it kind of a lateral extension, but we were pretty happy with the reception we got. So I think advertisers, they understood that we have the skill set, the technology, the data and kind of the mindset to help them do a lot more on the open web. So I think that was probably our most pleasant surprise in 2025 was just the reception we got to the rollout.

It's obvious when you put it in those terms, how big of an opportunity this is. So absolutely makes sense why you're going after it. Realize is now about a year old. What were you ultimately trying to unlock with this platform, particularly around performance marketing? And I guess performance is kind of an industry term. But in your mind, what does performance mean as it relates to Realize?

Yes. So yes, let me start with the second part of your question, which is what is performance. So for us, and I think in general, but you can identify a performance advertiser versus, say, a branding advertiser or somebody with other goals because they've got a very specific outcome in mind, whether that's getting somebody to buy a product, sign up for a service, give them your e-mail address that they can then market to you further from there. There's some outcome that they can measure. That's by far the most important thing.

But then second, they've got a very specific goal when it comes to that outcome. So a particular return on ad spend, a particular cost per acquisition, a particular cost per lead that they're targeting. So they can measure the outcome. They've got a specific goal. And then performance advertisers tend to be optimizers. So they tend to really focus on testing, trying different things, improving upon whatever that outcome is that they want. And that's kind of how you think of a performance advertiser.

And so in terms of what we are trying to unlock then, we're basically trying to help those advertisers reach consumers in the open Internet because at the end of the day, today, those advertisers are -- they're kind of limited. They're basically competing for space on Google and Meta are the big two, Amazon, if you're an e-commerce or retailer, but you're basically competing for space on what are increasingly competitive and crowded spaces. And we think we can help bring them to an audience that's in a way, underserved out in the open Internet and help them grow their business faster by reaching a whole new audience. So that's what we're really trying to unlock is we're trying to help them, help those advertisers to grow faster by reaching an audience that they have a hard time reaching today.

That's terrific. And when you talk about the different advertisers, these are performance marketers. Can you give us a little bit more about the complexion of your advertiser base?

Sure. Yes. And in fact, I'll talk about this in the context of one of the big things that we did in 2025 was we basically decide -- we looked really closely as we were rolling out Realize and we said, who works well on our platform? What advertisers find success. And we found kind of clusters of advertisers that we saw were being successful. We call them ideal customer profiles. So they're performance advertisers, the way we just defined it. But on top of that, they're in certain categories where it just happens to perform well on our network. So we focused our sales teams on those ideal customer profile verticals, and that was a big kind of focus for what we were doing in 2025 was getting our sales teams out there focused on those verticals.

So to your question about defining our base, those verticals tend to be e-commerce, that's our biggest one, finance and financial services, travel, direct-to-consumer products like a food service or something that goes direct-to-consumer, auto, travel, health. It's kind of some of the classic performance advertising categories, so not surprising. But those are the verticals that we found a lot of success with and the ones that our sales teams are now focusing on.

Awesome. So Realize is off to the races. You had a great 2025. Let's look forward to 2026. As we think about the year ahead, what are you specifically trying to accomplish with the Realize this year? And what should investors be watching to measure that progress?

Yes. I mean, ultimately, the goal of Realize for us is to reaccelerate our growth rate. So last year, we started the year guiding to 2%. We ended up at 7% growth year-over-year. Frankly, that's not where we want to be long term. We want to be a double-digit growth company. So that's what Realize is about is getting back to that double-digit growth. So the key metric to watch within that, so we released a metric beginning of last year that we call scaled advertisers. So a scaled advertiser is any advertiser that spends more than $100,000 with us in a 12-month period. And so that's an important measure for us because that, a, it makes up 85%, 86% of our revenue. So obviously, it's -- those are our key customers.

But on top of that, it's a good metric to track because anybody who spends $100,000 with you in a 12-month period, they're no longer testing. Like they have found some level of success with your platform or else they wouldn't be spending that much. So we measure that and we watch that internally because every time we get someone at that level, we know, okay, they found something that's working. So we released that.

And the metric, in particular, that we've released two metrics. One is the number of scaled advertisers and the second is the average revenue per scaled advertiser. I think the first one, the number of scaled advertisers, that's the one I would keep my eye on as an investor because that's the one that really says, are they growing their customer base? It's fuel for future growth because every time you get someone scaled to that level, I think you have a much better chance of growing them going forward. So I'd be watching the number of scaled advertisers.

Awesome. And getting back to that sustainable double-digit growth feels like these scaled advertisers are a key component of that recipe for success. Anything else that you guys are driving to get to a durable level of double-digit growth?

Yes. So I think there's really 3 things that we talk about internally and that we're focused on to help us get back to that double-digit growth rate. So first, I mentioned this a bit earlier, but we've got our sales teams now focused on ideal customer profile customers. So those are the -- and we've verticalized our sales team. So we have an auto vertical. We've got a finance vertical. We've got a health vertical. And I think what that does for us is it lets us have deeper conversations with those customers.

If you're always talking to auto advertisers, then you can talk the talk, you know who they are. We think ultimately that, that's going to help us grow the revenue faster with those advertisers and frankly, bring in more new ones because, again, if you know auto, then when you go out to talk to Ford Motor, if you've been talking to GM and Mazda and others, you understand their business, you understand what they're trying to do. So that's one thing, ideal customer profile and focusing on those verticals is one thing we're doing.

Second thing is, and this is an important one, we need to shift the brand perception of Taboola. So we need advertisers when we walk in the room to think of us as a full platform for performance advertising of any type. Not just native advertising, which is a form of performance advertising, but it's a small subsegment of it. We want them to think of us when we walk in the room as everything performance. And we're making progress. As we talk more to customers, they're getting -- like I said, we have great reception to that message. We just need more people to understand it.

And then third thing is the technology platform. So we've built a great AI-driven platform over time that takes that unique intent data that we have and helps match the ads to the consumers at those moments of intent in the open Internet, but we have to keep improving that. And in particular, we're focused on ways of making it really easy to use and also perform well for more and more specialized types of advertisers over time. So those are kind of the 3 areas of focus for us.

You touched a little bit on the go-to-market motion with these ideal customer profiles. Can you talk a little bit more about how you're empowering your sales team to go out and not just win new customers within each of these verticals, but actually scale existing spend with existing customers?

Yes. So I think -- and that's where the verticalization, I think, comes in. So if you -- we verticalized our sales team, so like I said, we can have deeper conversations with our customers in their industry specifically. I think then the next step of that is you get the case studies, you start to see the success like if Mazda is having a great success with us driving test drive, which is the type of thing that auto advertisers like to do with performance advertising. So they're seeing a lot of test drives coming out of campaigns with us.

Well, then you get the case study from them and you go talk to Ford Motor and you say, look, they're getting a ton of test drives, you should be working more with us on that. So if Ford is not working with us at all, that lets us have a better conversation about why they should work with us. If they are working with us, it gives us a good reason to go kind of work on growing their dollars that they spend with us. There's also a technology component to it, too. So I think we -- I don't remember if we've talked about it specifically, but we've released features on our platform now that also helps the sales teams.

So for instance, one is called predictive audiences. So what it does is it says, hey, Ford Motor, if you're spending $100,000 with us today and you're getting 1,000 test drives out of that, if you spent $150,000 with us, you'd get an extra 400 test drives. It predicts that for them. It just makes it a lot easier for our sales teams then to make the pitch as to why they should spend more over time. So there's also technology elements like that, that we're rolling out over time.

Fascinating. So not just driving clicks or downloads or other return on ad spend metrics, it seems very customized to the different verticals that you serve across the advertising base?

Exactly. Yes. It's a lot about understanding what they want and then figuring out how we can show them that we can deliver that.

Awesome. So let's zoom out a little bit. We talked a little bit about Realize. What are the other strategic initiatives that you have this year? And what would you like investors to understand about how you plan to execute against each of those?

Yes. To be honest with you, it really is about Realize for us right now. So we think that focusing on Realize is the right way to grow our business. And I mentioned -- so there's the go-to-market strategy that I think we've talked about quite a bit. There's the rebranding. So we are spending a lot more now on marketing of our brand. So getting out there at different conferences, making sure people are aware that we can do a lot more for them than just native advertising, which is how a lot of people still think of us. So we're spending more money there.

We're also doubling down on a lot of our investments on the technology side. So beyond just the things that we've talked about, for instance, we have AI features and AI tool called Abi, which is a -- it's a generative AI, LLM-based AI that can help do everything for an advertiser. It will set up your creatives for you, tell you how you should adjust your campaigns, help you figure out which bidding strategy you use. It's kind of an end-to-end solution. So I think bigger picture is really all about Realize. That's where we're focused, and that's where our attention is.

So you brought it up first. You said AI first. So I want to skip ahead to some of the questions on AI. So bringing all of this back to Taboola, what does the rise of AI specifically mean for your business? And how are you thinking about areas like search and traffic -- search traffic and discovery and the impact that AI will have on those businesses?

Yes. So I think we -- and I'm sure you hear this from everybody, but we view AI as an opportunity more so than a threat. The reason we say that is if you think about -- and I know what a lot of the markets are worried about when it comes to SaaS and other things right now is that AI is going to allow companies to kind of rebuild whole SaaS platforms without having to buy it from somebody or -- so SaaS can be displaced.

I think in our case, the way I think about it is AI is going to help us build a much more robust platform, do a lot more. And maybe it could help a start-up or some other competitor build our platform. But to me, that's like if AI could help someone build the most amazing sports car in the world, if you don't have fuel to run it, it just doesn't matter. And the fuel that runs our platform is the data -- the intent data that we gather by having distribution on thousands of premium publishers around the web. Without that data and knowing what works, what ads work in which context for which consumers, you can build the best platform in the world, and it just won't perform.

So I think we have a very unique positioning in that we've got the data, we've got the distribution that's going to allow our tech platform to work, and we're going to use AI to make that tech platform dramatically better. I mean, I hear from our engineering teams how they think our 600, 700 engineers can become like 7,000, 10,000 engineers because of AI. So that's exciting. And again, we've got the fuel to then make that work.

Now the important within that is that we need the distribution to get that data. And I think we believe that we are our total base of publishers is only about 5% to 10% search traffic. So what that means is that most of the traffic to our publishers comes direct. It doesn't come through search. So I think the -- our advertiser -- or sorry, publisher base because of the premium publishers we work with are less exposed to the potential threat of ChatGPT or somebody taking away all of their traffic. In fact, we've seen the traffic to our publishers grow year-over-year even with LLMs coming in.

So I think we feel like we're in a very good position where we've got the distribution, that distribution seems fairly secure and safe. And then that drives the data that allows us to really make our platform perform. And then we'll use AI to make our platform that much better. So that's why we think of it as more of an opportunity than a threat at this point.

Yes. It's hugely self-reinforcing between the data, the distribution. And really, the distribution, you're working with premium publishers who represent a destination for so many of their users, which substantially derisks a lot of that AI threat.

Yes. And I think one of the signals that we've talked about as to how sticky our publishers are is that over 1/3 of our traffic is in-app, meaning it's like ESPN's app or the Yahoo! Finance app or CNBC's app. So people who come to an app, that just proves that it's a much stickier audience because it's not coming through search or coming through -- it's not fly by traffic. It's somebody who comes there regularly.

So let's talk a little bit about the financials. Can you walk us through the 2026 guidance that you recently published as well as what you're seeing underneath it in terms of the customer verticals, channel mix, broader ad spend trends?

Sure. Yes. So we're guiding this year. I mentioned that last year, we started guiding at 2%. We ended up growing 7%. We did about 30% EBITDA margins. And we've been saying for a while, as you think about 2026, start by thinking about our 2025 results as a basis. So we're guiding to 7% year-over-year growth on ex-TAC. We're guiding to 30% EBITDA margins. We've said that we think we should convert 60% to 70% of our EBITDA into free cash flow. So -- and then over time, what we want to do is get that 7% growth rate up to double digits.

For now, what we've built into our guide is basically what we're seeing on the platform today. And the key things that drive kind of the growth rate of our platform is new customers, how many new customers are we bringing on and how much revenue do they bring in, which the key to that is how successful are they? Like you can go out and get a test budget, but you got to make them successful to get more budget. So it's new customers and their success rate. And then it's what we call NDR, net dollar retention. So it's for the customers that we already had, how do they grow over time. Those are the 2 key metrics.

And so what we built into our guide this year is kind of what we've been seeing as we've -- we talked in Q3 of 2025 about seeing an inflection point. So we built that into our model, but we haven't built kind of upside from some of the key initiatives that I talked about earlier. So the verticalization and the focus on ideal customer profiles, hopefully, our brand gets stronger over time, and it gets easier to get new customers in and that technology to be rolled out, should drive better performance, too. But we kind of didn't build those things in, that's option value. We built in what we're seeing today.

Right. That makes sense, appropriately conservative. And then you talked a little bit about where you're investing. Let's talk a little bit about capital allocation more broadly. You've historically focused on being "cash neutral." How should investors think about capital priorities this year, including buybacks?

Yes. So we've been pretty consistent with this over the last 18 months, basically. I think we still think the best use of our capital is for share buybacks. We still expect to be generating a large amount of cash flow. We're guiding to over $230 million of adjusted EBITDA. And if we convert 60% of that into free cash flow, that's a decent amount of free cash flow that we have. And we do think that the best use of that is for share buybacks. We have said all along that we're always looking at the M&A markets.

And if we found something that we find interesting, we would look at it, but we expect those to be smaller, more tuck-in types of things where we're acquiring capability that we particularly like and feel we need or maybe a small customer base somewhere that we think it's interesting, but it's likely to be small and tuck-in. And therefore, we expect most of our capital to continue to go towards buybacks. We still have $180 million left under our current authorization. We've said that we expect the majority of our free cash flow this year to continue to go to those buybacks.

Okay. Awesome. Last question for me before we open it up to the floor. And just to wrap things up, if we're sitting here in 3 to 5 years from now and Taboola has really succeeded in what you're trying to build with Realize and some of these other strategic initiatives, what will most like -- what will the company look like at that point in time versus where it is today? And where do you think investors will say you got it right?

Yes. So I think, obviously, it comes down to Realize and the success of making this transition to be performance for the open Internet. So if we're truly successful, what's going to happen is you'll talk to an advertiser and you'll say, "Hey, where do you go -- performance advertiser in particular, where do you go to spend your dollars? And why do you go to those places?" And what they'll say is -- let's assume they're not a retailer for a moment, they're a Ford motor. I keep coming back to them for some reason today. But they're Ford motor. They'll say, "Oh, well, we go to Google for search, we go to Meta for social, we go to Taboola to reach the open Internet." Like that's our -- that's what we want to be.

We want to be the channel for advertisers to reach consumers in the open Internet. We want to be the third leg of the stool or if you include, if you're a retailer, the fourth leg of the stool because then Amazon is a key part of their strategy as well. And we think that's important for advertisers because, like I said, today, they're kind of trapped into these very expensive, very crowded walled gardens. And if we can become that third leg, it's valuable to them, obviously valuable to us in growing our business. And frankly, it's valuable to publishers as well as they try and grow their business.

Awesome. Well, Stephen, thank you for your time. I'd like to open the floor to any Q&A that we have in the room. I see a hand over here.

I was hoping like if you look more broadly at ad tech in general, what is it that you think investors misunderstand about the space, competitive advantage, durability of the business model, terminal values, clearly, something is being misunderstood because when you talk through your numbers and you talk through your growth, there's something there that you believe in, in 3 to 5 years, if you are that third leg of the stool, then it's a wonderful business, but something is clearly not clicking right now. And you might not want to talk about competitors, but maybe Taboola specifically.

Yes. Yes, I'll focus on Taboola specifically because I don't want to talk for the industry as a whole. But I think for Taboola, I certainly think that what's -- there's probably two things that are being missed. So one is what I talked about. I think right now, there's a lot of concern out there about LLMs being able to just rebuild tech platforms from scratch. But I think that's where I don't think people are fully understanding how important the data is that we have and how important the distribution is that we have to get that data, being on thousands of premium publisher websites globally is really the only way to get that.

So if you think of it, if you're a, let's say, a start-up company and you're coming in to try and disrupt what Taboola does, you may be able to build a great tech platform, thanks to AI and everything else. But then you go to one of our publishers and you say, give me your business and they say, well, how much are you going to pay me? And the answer is, unless you're really, really well funded and want to lose a lot of money, the answer is "I can't pay you that much." So there's a big chicken and the egg problem here that comes that I think is being misunderstood or at least not fully valued. So I think that's one thing.

I think the second thing is, I think there is some fear that LLMs disrupt the open Internet as a whole, right, that the open Internet -- traffic to the open Internet goes away. We don't believe that. And in fact, we are actually seeing the signs. Like I said, our base is actually growing year-over-year. Why is that? Well, my belief is that it's because the sites that are getting hit are the long tail, the ones who don't have a brand and don't have a regular user base. So if you're a Boston Celtics blogger who has a little blog, which is popular, but frankly, doesn't have a brand per se, you can -- people can ask questions of an LLM and maybe they stop coming to your site.

But at some point, people do want to go watch the video from the Celtics game last night. They want to read from commentators they trust. So they go to ESPN. And I think brands like ESPN and CBS Sports and CNBC and Yahoo! Finance and all the brands we work with, I think they have a long-term place. I mean, frankly, without them, what is -- where is LLM even going to get their content. So I think there's a long-term place for that, and I think that's being misinterpreted right now, too. There's a lot of fear, and I think that people will figure out over time that it's overdone.

Maybe just one follow-up question because I do think that resonates with me on the LLM and AI risk. But what about just broader competition throughout the industry? I think you're seeing that kind of trickle down with players like Amazon, for instance, getting more aggressive on the ad tech side. How do you see that playing out longer term?

Well, I think -- so obviously, Amazon has disrupted the DSP space. I think that's where you're seeing it big time right now. I think, though, for us, the way we think about it is if you are not adding a lot of value to whatever your kind of customer base is, both supply and demand, then you're probably exposed. I think we've said for a long time that if you're on only one side or the other, you may have a challenge. Like if you're an SSP who only works with publishers and you don't have direct demand, that's disruptible. If you're a DSP who only works with advertisers, you don't have direct access to supply, that's disruptible. And I think, especially when the person with the most supply comes in, which is Amazon, right?

So I think we believe that, first, you want to be end-to-end. So you want to have relationships on both sides, which we do. About 90% of our revenue comes from direct advertisers and almost all of our distribution is direct relationships with publishers. So you want to be 2-sided. And then you got to make sure you add a lot of value in between. And I think what's unique about Taboola and the value we add is that we know intent on the open Internet. So just like Google knows intent on their platform and Facebook knows intent on their platform, we know intent in the open Internet, and that's a nontrivial challenge, like to have enough data to predict that this ad in front of this consumer in this particular context on the open Internet is going to work is a challenge.

It's a bigger challenge, frankly, than search. It's a bigger challenge probably in many ways than social. That's the value add we bring. So I think the ones who are most at risk of being disrupted are people who don't have that value add. They're either one-sided and therefore, can't add enough value on both sides or maybe they're 2-sided, but they don't really have real value add in between. So I think that's where we think we're in a fairly good position because of that value add.

May I ask you about how you feel whether your data is protected. How do you develop your AI tools? Do you use Gemini OpenAI? And if you use this frontier model, do you feel that actually what you possess is really secure?

Yes. No, it's a good question. I mean, wow, things are changing so fast that I have a feeling any answer I gave you today would be different by next week. What I can tell you is we've got 700 engineers globally. Our biggest group is in Israel. By the way, pray for everyone in Israel right now because it's a little bit noisy there. But our biggest group is in Israel. We've got a big group in L.A., Taiwan. We have Eastern Europe. So we've got a very large engineering team, and they've been cutting edge on AI for a long time. And I think what I trust them to do is figure out the best tools at any given time.

And so when I talk to our Senior Vice President of Engineering, he gets me excited about how we can go from being kind of the small guy in many ways relative to the big walled gardens having 600, 700 engineers, leverage them using AI and act like we've got 20,000 engineers. And now we can really develop things that were just pipe dreams a year ago. So the answer is to kind of avoid your question, it changes every day as to which tools we use, but we've got a really strong engineering team who knows how to use the tools and are building ways to leverage the engineering team and make them way more productive than they've ever been able to be.

But using third-party LLMs, is it dangerous for your business? Or do you feel the risk of exposing yourself to the fact that if you use the -- either LLM or even the agent provided by OpenAI, for example, whether it exposes you to the risk of losing some of your data?

Well, yes, so -- and there's 2 sides to this. When you first asked your question, I thought you were talking about engineering tools like Cloud to help you code or other tools like that. I think most of those services expect themselves to be tools forever or at least for a long time. So I don't think there's -- they have to help you avoid that exposure or else they won't have a business for very long. So when it comes to engineering tools and making your teams more productive, I'm not as concerned. But I think your question also was related to using LLMs within our products and to do things.

I think we watch that. So we obviously are very cognizant of what's the licensing on this, what are they allowed to do? What are we allowed to do? And we won't use tools that we're giving away data for. So like we know that our data is our secret sauce, and it's what makes us valuable. So we're very careful to not use tools where we're exposing that data to the world. We only use tools where they will allow us to run instances on our own servers, our own infrastructure, firewalled from the rest of the world where they can't get that information. So we're very careful about that.

Okay. Thank you very much, Stephen.

All right. Thank you. Appreciate it.

Good day, and thank you for standing by. Welcome to Taboola's Fourth Quarter and Full Year 2025 Earnings Conference Call. [Operator Instructions] Please be advised that today's conference is being recorded.

I would now like to hand the conference over to your first speaker today, Aadam Anwar, Head of Investor Relations. Please go ahead.

Thank you, and good morning, everyone, and welcome to Taboola's Fourth Quarter and Full Year 2025 Earnings Conference Call. I'm here with Adam Singolda, Taboola's Founder and CEO; and Steve Walker, Taboola's CFO.

The company issued earnings materials today before the market, and they are available in the Investors section of Taboola's website. Now I'll quickly cover the safe harbor. Certain statements today, including our expectations for future periods, are forward-looking statements. They are not facts and are subject to material risks and uncertainties described in our SEC filings. These statements are based on currently available information, and we undertake no duty to update them, except as required by law.

Today's discussion is also subject to the forward-looking statement limitations in the earnings press release. Future events could differ materially and adversely from those anticipated. During this call, we will use terms defined in the earnings release and refer to non-GAAP financial measures. For definitions and reconciliations to GAAP, please refer to the non-GAAP tables in the earnings release posted on our website.

With that, I'll turn the call over to Adam.

Thanks, Aadam. Good morning, everyone, and thank you for joining us today. We're closing up 2025 with another strong quarter, exceeding the high end of our guidance across our key metrics. The year has been defined by disciplined execution and more importantly, we're seeing clear early signs of acceleration in the growth of the business from our new advertising platform, Realize.

In 2025, we repurchased 77 million shares for a total of $254 million, reducing our share count by roughly 18%, while continuing to invest in R&D to support our long-term growth ambitions. Before getting into the details, let me remind you who we are. Taboola is one of the largest performance advertising companies outside of search and social, focused on the open web.

Every day, billions of consumers read, watch and engage with trusted publishers and communities across the open web. Similar to how Google and Meta understand intent within their own platforms, Taboola understands intent across the open web and turns it into measurable outcomes for advertisers. When someone reads about the Knicks, plans a vacation or checks the latest news on their favorite local site, we transform that moment of interest into measurable results for advertisers. That scale, that proprietary intent data and the AI-driven conversion machine we've built, that is Taboola.

Turning to our results. In 2025, ex-TAC gross profit reached $714 million, up 7% year-over-year, and adjusted EBITDA grew 7% to $216 million. We began the year guiding for 2% and exited the year at 7%, a clear acceleration, which I'm happy about, while I believe double-digit growth is the right long-term pace for this business. We're not there yet, but our 2025 performance gives us the confidence we're going in the right direction. We also generated $163 million in free cash flow, up 10% year-over-year, representing approximately 76% conversion from adjusted EBITDA.

Looking ahead, we expect 7% ex-TAC gross profit growth and 30% adjusted EBITDA margins while continuing to invest in accelerating our growth rates and continuing our primary use of cash to aggressively repurchase shares. In 2025, Realize, our advertising platform, helped increase the number of scaled advertisers and grow the budgets we manage for them. In 2025, scaled advertisers grew 6% with an average revenue per scaled advertisers up 2%.

These results are reflected in the financial performance I shared earlier. A strong example is personal finance, one of our ideal customer profiles. Advertisers such as NerdWallet, Motley Fool, and QuinStreet adopted Realize and leveraged newer capabilities like predictive audiences and format diversification. As a result, they grew meaningfully beyond their historical spend levels with some becoming top advertisers at Taboola.

When I think about what will continue accelerating Taboola's growth, I am laser-focused on improving retention rates and increasing spend over time. While many things can help, this is the most important one. Examples like these encouraging and reinforce that our strategy is working. As we look ahead, we are concentrating on these 3 priorities:

First, investing in our technology to continue to advance Realize as we continue to expand our strategy to become the leading performance advertising company outside of search and social. We're investing heavily in AI-driven optimization, predictive targeting, onboarding automation and stronger measurement and attribution to make the platform even more intelligent and easier to adopt while directing budgets towards the best performing opportunities. While I think we're making good progress, there's a lot more for us to do here, and our R&D team is hard at work rolling out capabilities that advertisers are asking us to further drive advertiser success.

Second, we restructured our sales organization around ideal customer profile, where we are seeing stronger retention and spend growth over time. The advertiser outcomes we delivered in 2025 are giving us clear signals on which advertisers to prioritize, how to reach them and what success on Realize should look like. To further support these efforts, we recently welcomed Krishan Bhatia as our new Chief Business Officer, overseeing revenue and partnerships and bringing additional focus and expertise to supercharge advertiser, agency and publisher relationships to accelerate growth.

Keeping with the same example I mentioned earlier, in 2025, we generated $120 million in personal finance revenue within a $15 billion U.S. market. Today, we capture only 1% to 10% of advertisers total spend, which underscores the significant runway ahead as we deepen those relationships. At the same time, we are prioritizing new advertisers similar to the ones already succeeding on our platform and entering those conversations with a clear understanding of their goals and what performance they should expect from Realize. By focusing on the right advertisers, not just volume, we're strengthening partnerships, expanding wallet share and positioning Taboola as a core long-term growth channel for advertisers.

Lastly, on brand and perception. Since launching Realize one year ago, we've made meaningful progress in how advertisers view Taboola. As advertisers see clear results and expand their budgets with us, we're building trust and steadily positioning ourselves as a platform advertisers should test and scale beyond search and social. There is still work ahead, but Realize is proving to be a strong engine not only for performance, but also for long-term brand credibility.

As we think about our partners and the open web in the context of AI, this is one of Taboola's greatest structural advantages. AI is a commodity. Anyone can download an open source of Llama and get going. AI can replicate features. It can improve interfaces. It can even outperform some raw models we developed. But that alone just doesn't matter. Without proprietary data and distribution, it is a very powerful engine with no fuel.

Our data is our fuel, and it is unique to Taboola. Hundreds of millions of times every year, people across our network make decisions to buy, subscribe or take action. That creates a very rare form of performance-driven intent data that directly determines advertisers' outcome. Think of it as a secret language of intent that exists only because of our deep integrations across the open web and our singular focus on performance advertisers.

Without these signals, advertisers cannot effectively optimize, scale or generate strong returns on investments. We get this data by having code on page integrated across 14,000 publisher properties such as ESPN, Yahoo, USA Today, The Independent and many others, giving us first-party access to more than 600 million daily users. Those direct relationships built over many years generate real-time intent signals at massive scale. When I look at our partners, what stands out is the strength of their brands, the trust and communities they've built over many years.

Users go directly to those, whether through their websites or their dedicated apps. As a result, they have little to no reliance on search traffic, while direct traffic continues to grow. These dynamics keep our company-wide exposure to search in the single-digit percentages with about 1/3 of our supply coming from in-app usage. In an AI-driven world, 2 assets ultimately matter most, proprietary data and distribution, and we have both.

In summary, 2025 was not just about beating the numbers, but further validation that our strategy is working. We executed with discipline, accelerated the business, returned significant capital to shareholders and invested heavily in the platform shaping our future. Realize is delivering the type of results we want to see, making new and existing advertisers successful while changing how the market sees Taboola.

We are still early, but we're operating with greater clarity and urgency than ever. Our mission remains to help performance advertisers grow, help publishers win and build the leading performance advertising company beyond search and social. As more players compete for advertising budgets, they will all need a trusted friend and Taboola is a great friend.

With that, I'll hand it over to Steve.

Thanks, Adam, and good morning, everyone. We are pleased to close out the year on a strong note. In the fourth quarter, we continued to build on the momentum we generated throughout the year, delivering results that exceeded the high end of our guidance across our key metrics.

Revenues in the fourth quarter grew 6% to $522.3 million and for the full year increased 8% to $1.91 billion. One of our key priorities this year was expanding advertiser budgets and with the rollout of Realize, our performance advertising platform and the introduction of new embedded features, we were able to successfully execute on that objective. This momentum was reflected in our scaled advertiser metrics in the fourth quarter with a 3% increase in the number of scaled advertisers and a 2% increase in average revenue per scaled advertiser.

We also enjoyed strong growth from non-scaled advertisers during the quarter, which contributed about 1% to our year-over-year growth. This indicates that we had a large number of advertisers testing Realize for the first time even if we have not had a chance to scale them as of yet. For the year, scaled advertisers grew 6% and the average revenue per scaled advertiser grew 2%. Realize continued to improve retention and increase ad spend among existing advertisers compared to the same period in the previous year.

As I've noted in prior quarters, we're particularly encouraged by growth in the number of scaled advertisers as they continue to be an important driver of future growth. Ex-TAC gross profit in the fourth quarter was $212.8 million, representing margins of approximately 41%. The fourth quarter results were flat year-over-year as expected due to the lapping of a challenging comparison with a strong Q4 2024. For the full year, ex-TAC gross profit grew 7% to $713.5 million.

This growth was largely driven by the scaling of Realize, which drove growth in advertiser spend as well as continued strong performance from Taboola News. Gross profit for the quarter reached $175.6 million, with full year gross profit totaling $569.5 million. In addition to growth in ex-TAC gross profit, this performance was driven by lower depreciation expenses on our servers following a reassessment of their useful lives as well as tax efficiencies, both of which offset higher hosting and data costs required to support the growth and scaling of our business.

In the fourth quarter, net income was $50.1 million with non-GAAP net income coming in at $79.1 million. For the full year, net income was $42.3 million with non-GAAP net income coming in at $168.6 million. Adjusted EBITDA for the quarter was $86.1 million. For the full year, adjusted EBITDA was $215.5 million, representing a margin of 30%. This reflects continued discipline in expense management while maintaining targeted investments to support long-term growth.

Foreign exchange was a meaningful headwind in the quarter. On a constant currency basis, Q4 ex-TAC gross profit saw a tailwind of approximately $4 million, while operating expenses saw a headwind of approximately $7 million, primarily reflecting the strength of the Israeli shekel, where we have a significant employee and cost base. In total, FX represented roughly a $3.5 million headwind to Q4 EBITDA and about $11 million for the full year. Without this FX headwind, our full year adjusted EBITDA would have been $226.3 million, which would have represented an EBITDA margin of 31.7%.

In terms of cash generation, we had $59.7 million in operating cash flow in the fourth quarter and free cash flow of $46.9 million. For the full year, operating cash flow amounted to $208.4 million and free cash flow was $163.4 million, representing a 76% conversion from adjusted EBITDA. On average, our free cash flow conversion from adjusted EBITDA has remained above 70% over the last 12 consecutive quarters.

As a reminder, last quarter, we indicated that we now believe we can sustainably convert free cash flow at a 60% to 70% rate over any typical 4-quarter period. That is an increase from our prior expectations of 50% to 60%. Capital expenditures in 2025 included internally developed software that were capitalized during the year, and we expect these strategic investments to continue into 2026. These investments were primarily driven by 3 initiatives: continued development of Realize, investment in new publisher-focused product capabilities and investments in our e-commerce platform.

Turning to the balance sheet. We remain in a strong financial position. We ended the fourth quarter with a net cash balance of $18.6 million. Cash and cash equivalents totaled $120.9 million, which more than offset our long-term debt of $102.3 million. Early in 2025, we secured a $270 million revolving credit facility, which enabled us to fully repay our prior term loan while maintaining approximately $168 million of available liquidity as of December 31.

The facility also reduced interest expense by $1.1 million in the fourth quarter and $4.8 million for the year. We remain focused on disciplined capital allocation, prioritizing R&D investments while returning excess capital to shareholders via share repurchases. In the fourth quarter, we repurchased approximately 18.6 million shares at an average price of $3.78 for a total consideration of $70.5 million. For the full year, we repurchased 76.9 million shares at an average price of $3.30, which represented total repurchases of over $250 million.

In 2025, we bought back about 18% of our outstanding shares net of issuances. This reduced our total shares outstanding to approximately 276 million at the end of 2025 from about 337 million at the end of 2024. Since the inception of our share repurchase program in 2023, we have repurchased a total of 110.4 million shares at an average price of $3.49 for a total consideration of $383.5 million. We currently have approximately $180 million remaining in our authorization and intend to continue to use a majority of our free cash flow to repurchase shares.

Moving to guidance. For the first quarter of 2026, we expect revenues to be between $444 million and $462 million, gross profit to be between $119 million and $125 million, ex-TAC gross profit to be $158 million to $164 million, adjusted EBITDA to range from $20 million to $26 million and non-GAAP net income to be from negative $1 million to positive $7 million.

For the full year, we expect revenues to be between $1.99 billion and $2.05 billion, gross profit to be between $601 million and $621 million, ex-TAC gross profit to be $753 million to $774 million, adjusted EBITDA to be $222 million to $236 million and non-GAAP net income to be $165 million to $191 million.

I would note that our adjusted EBITDA guidance reflects a forecasted headwind from foreign exchange rates of approximately $11 million in operating expenses, partially offset by ex-TAC tailwinds. Without this headwind from foreign exchange, adjusted EBITDA margins would have been over 31%.

In summary, Q4 results exceeded the high end of our guidance across our key metrics, reflecting strong execution and continued momentum in the business. We are building on the traction we've seen with Realize and are focused on accelerating growth as our initiatives gain more traction this year. While we remain disciplined in our approach, the progress to date reinforces our confidence in our ability to return to sustainable double-digit growth over time.

With that, let's move to Q&A. Operator, can you please open the line for questions?

[Operator Instructions] Our first question comes from Barton Crockett from Rosenblatt.

One thing I was curious about, you didn't really address it in the commentary, but -- and I realize maybe this means you don't -- it's not a KPI, but there was a substantial variance in your revenues versus where you were guiding for the quarter. And I was just wondering if you could talk through what that variance was, why it happened and how meaningful that is?

Sure. I can take that. So I think very simply, it was revenue mix -- or mix of business. So we had more business in kind of some of our higher-margin parts of our business and less revenue in some of our lower margin. So ultimately, it was just mix of business. Obviously, for us, gross revenue is not the key metric. Ex-TAC is the key metric because that's what we keep after we pay publishers.

You've probably heard me say a bunch of times in the past that we can grow gross revenue by doing bad business, signing up a bad publisher deal or doing something that doesn't drive ex-TAC, and that's not helpful. So what we care about is ex-TAC. So we're obviously happy that we had the beat on ex-TAC, which is really what we focus on. The rest of it was just mix of business.

Okay. And then you guys gave the commentary about the growth in non-scaled advertisers suggesting some success with the real life initiatives to grow penetration in other elements of the page beyond bottom of page, which sounds encouraging, but your guidance suggests kind of a steady revenue trajectory versus acceleration.

I was wondering if you could talk through that kind of disconnect. I mean, how optimistic are you really that this can bring enough new business in to move the top line? And why isn't that reflected in the guidance that you gave?

Yes. So I think ultimately, our guidance philosophy as a company is always to be relatively conservative, so we don't want to get ahead of ourselves. So what our guide basically implies right now for 2026 is what we're seeing from Realize at this point in time. So we've obviously seen good progress with Realize over the course of last year. We started last year guiding at 2%. We ended the year at 7% growth. We're now midpoint of our guide for 2026 is at 7%. That's because that's basically what we're seeing from Realize today.

We do have initiatives that we think will help improve that over time, and Adam can probably talk to a few of those initiatives that he thinks will drive growth this year, but those are not factored into the guide yet. So for now, what we're factoring in the guide is exactly what we're seeing today. Adam, did you want to talk about some of the things you think can grow?

I think in general, we're encouraged by seeing our investments in Realize at the center of our strategy progressing, right? So the 3 things I mentioned. The first one is just focusing on our technology side. And we're seeing better retention for new advertisers, which is probably what we want to see the most. And we're seeing growth in spend over time. The second thing, which -- again, and that results in scaled advertisers, which are growing and all those things are positive signs that we're progressing in our strategy and its results in our numbers, as you can see from 2025.

The second thing, I just came back from Bangkok, from Madrid, from Chicago, spending time with our 600 sellers. It's really incredible to spend time with our people and seeing that when you sell to the right clients, we call those ideal customer profile, we're seeing essentially -- we have what it takes. The chances for -- I'll give you an example, the chances for a financial advertiser to succeed with us is not too different if they were to spend with Meta, which is incredible because it means that there's so much growth for us within our existing markets that we're going after. So the second thing is just sales focus and going after the ones that we know chances for success are much higher.

And the third one is continue to invest in our brand. I think it's quite, for me, always encouraging to see how many advertisers don't even know Taboola is out there. So there's so many great advertisers that should try Taboola that will succeed with us or they have a good chance to succeed with us. And as part of that, I think, continue to invest in our brand perception and our brand in general will continue to help us attract new advertisers to try Taboola and succeed with us. So all those 3 things make us encouraged.

Our next question comes from Matthew Condon from Citizens Bank.

Adam, you talked about making incremental investments just behind the product features and Realize. Can you maybe just dig into some of those and what we should expect from a product perspective in 2026? I think I was wondering if you could just break down a little bit more as we look at Realize and just like how much is coming from existing advertisers and you tapping into incremental budgets there versus bringing in new clients onto the Taboola platform?

Sure. I'll let Steve speak about the numbers. But the biggest investment we're making, and I think the biggest opportunity for Realize, and we'll share more throughout the year. So I want to let the team kind of bring this to market in a more detailed way. But in general, what I think we have the biggest opportunity is making it more automatic and simpler for advertisers to be successful. If you look at the amount of permutations that exist when you buy from any channel, by the way, whether that's Google, Meta, Taboola and others, it's complicated to succeed as a performance advertiser.

So even right now, where Taboola is with Realize, I think we made a tremendous progress in terms of making advertisers successful. In my vision, I really want anyone that has a chance like that should succeed with Taboola to almost automatically succeed with Taboola. So I think with -- in a world of AI, where we have so much unique intent data, and we have so many thousands of advertisers that are already doing well with Taboola generating $2 billion of conversions a year.

On the other side of it, I hope that Realize is a platform that if you should succeed with us, then chances are you will succeed with us, and that will be more and more automatic. And then our good people that we have at the company can spend more of their time on strategy and being creative and going out there and help attract more new advertisers. So again, to me, the biggest thing that we'll see from Realize later will be more about automation and making it even easier for those who should succeed with us to drive success.

And then to the second part of your question about whether or not growth is going to come from bringing new advertisers to the platform versus growing our existing. And I'll talk about this in the context of our scaled advertiser metrics that we release. What I would say is the precise mix is always hard to predict because as I've talked about in the past, as we bring on more advertisers and we scale them and they get into that scaled level of performance with us, they do drag down the average.

So as the number grows, the average gets dragged down because usually, when we initially scale an advertiser, it's at the low end and then we grow them over time. So the exact mix is hard to predict. But in general, what I'd say is we always expect to grow the number of scaled advertisers. That's the fuel for our growth. And so I would think that a larger portion of our growth comes from growing the number and bringing more new advertisers to the platform. But we should see some growth in the average revenue per scaled advertiser over time as well. So I'd say it will come from a bit of both, generally speaking, probably more from the number. And then over time, we'll grow the average as well.

Our next question comes from the line of Laura Martin from Needham.

My first one is on generative AI. So are you -- I'm interested in whether -- how much your traffic was down in the fourth quarter and what the mix was and whether you think that, that -- I think Wall Street thinks that's the first step in Agentic holding on to attention and not allowing people to go to the open web. So can you talk about why the open web survives generative AI? I think that's my first question.

And then my second question is about Realize. One of our goals, I think your goals in Realize was to attract display budgets, which are quite a bit larger than native budgets. But I'm interested in whether Realize is actually -- are you seeing that happen that you're getting new types of advertising rather than just staying in the narrow native advertising bucket. Those are my two.

So I'll pick up the first one. So on the open web, we're basically -- I think we have a very -- like a structural advantage in where we sit in the open web. So one, I think that we're seeing traffic going up. We're seeing search traffic going down, but overall, through primarily direct traffic to publishers and then just onboarding more publishers, traffic is overall going up. And the exposure we have to search traffic, which I think is the main risk that investors are tracking for us, it's in the single digit. And a lot of it is because we work with massive platforms like Microsoft and Yahoo and Apple News, 1/3 of our traffic is in-app.

So overall, our exposure is low, and we're seeing direct traffic going up. And I think in general, what's going to happen is publishers that have trust, that have good communities around them will continue to be important local news, sports rights, news, they'll continue to get a lot of momentum and attention from consumers. I can also tell you, AI engines, what we're seeing is what they crawl on the web as a proxy for what consumers are asking, a lot of it -- a big chunk of what consumers are talking to AI is about the last 24 hours news. People want to know what's going on. That's -- so AI really needs that content and the open web is where content exists.

So I think that for trusted publishers, for bigger publishers, which is most of our business, there's a very bright future. And the second thing is that when I imagine AI being adopted by those publishers, as you know, we have a product called DeeperDive, which is essentially bringing ChatGPT type technology to those bigger publishers so that consumers can convert, can talk to publishers. We can -- if you go to use it today, you can check it out. I think there's a big kind of ARPU growth, a significant revenue generation opportunity for publishers when they actually adopt AI on their own sites. So the risk, I think, is more on the smaller sites, which we don't have exposure to or for those who are very dependent on search, also not publishers that we work with. So I think there's a very bright future for the trusted publishers and especially when they adopt AI in a bigger way.

And then to your second question, Laura, about are we seeing new types of advertisers coming on to the platform. I'll talk about this in the context of the 3 growth drivers that Adam mentioned earlier. So he said, we're focusing on ICPs. We're investing in our brand to change perception of who we are as a company, and then we're investing in tech to make advertisers more successful. I think today, that focus on ICP means we're bringing more of similar types of advertisers. So what we've done is we've got our sales teams focused on finance advertisers, travel advertisers, auto advertisers, e-commerce advertisers, the ones that we know are working well on our platform today.

So today, our growth in advertisers is coming more from that focus on ICPs and getting more similar types of advertisers to what we have. But what we expect over time is that as we get our brand perception shifted a bit like getting out of the -- we're a native company and into the -- we're a performance platform type of mindset. And as our tech continues to develop and we're able to target more and more granularly on our platform, we do expect that we will expand the types of advertisers. So more types of advertisers will become ICPs, and we'll start focusing on selling to them. So I'd say today, more of it is more advertisers of a similar type to what we have today. And then over time, I expect more different types of advertisers to start coming on.

Our next question comes from the line of Tyler DiMatteo from BTIG.

I wanted to start in terms of 2026, Steve, as you think about the advertising market this year and some of the one-off events, kind of FIFA, et cetera, is that baked into the guide? Is that -- I guess, what level of visibility do you have into something like that today? And kind of when would that start flowing through?

And then my second question for Adam, on Realize and the developments and just I'm thinking about this in the context of the investment cycle for that, kind of where do we stand in terms of the investments in the platform, the technology, et cetera? Are we going to see multiple iterations from here? Is everything largely ironed out? Those are my two.

I can start with the second one and then -- so one, I think we're all in. So this is -- we're laser-focused on Realize. I think that, like I mentioned earlier, if you just look at the market that we're selling into, the performance advertisers that we're going after with the technology we have now, I think we have what it takes to grow, and we spoke about seeing an inflection point into double-digit growth. And I believe in our strategy, the market and we have what it takes. Saying that, we're early in our cycle in terms of investment, in terms of there's so much more that we're going to reveal later this year and in years to come.

When you compare Taboola, Realize to Meta, when you compare it to Google, to PMax to some of the platforms out there that are serving 10 million advertisers when we serve 15,000 to 20,000, there's so much more that we want to do and intend to do. So I'll break to 2 parts. The first is that I think we have what it takes to continue to grow and to generate 30% EBITDA within that growth rate and convert 60%, 70% of that to free cash flow and use most of it to repurchase shares, which we think is a great deal for the company. And the second thing is that most of our investment, which is significant, is on technology as a technology company, and we're going to reveal a lot later in the year. And like I mentioned earlier, to me, something that's very exciting is can we make it so much easier for those who should succeed with us and become scaled advertisers to actually become ones, and that's later in the year.

Then to your first question about are the kind of big events that are happening this year factored into our guidance. The quick simple answer to that is yes. The way they're factored in, just to get into a little bit more detail, is like the big events this year are the Olympics, World Cup, midterm elections. Those things are factored in. For us, though, interestingly, it's more of a traffic driver than it is an advertising revenue driver. So if you think about the events, World Cup and Olympics tend to be big sports traffic drivers, and we're -- we have Yahoo Sports, we have CBS Sports, we have ESPN. We have, I think, something like 8 of the 10 top sports sites in the U.S., and we have similar coverage globally.

So it is a great traffic driver for us. Our advertisers, though tend to be always on performance advertisers more so than event-driven advertisers. So it will drive more traffic, which is more impressions and give an opportunity to drive more revenue from our advertisers, but it's not like a display network where maybe they've got event-driven advertisers. Same thing with elections. I think we've talked about this in the past. Elections drive big ad budgets, but a lot of that is branding campaigns for the candidates. We do get some things like fundraising campaigns where the -- it's a direct response trying to get somebody to donate. But we don't get a lot of incremental revenue in terms of the advertising side. But again, it drives eyeballs and drives views, and that's what's factored into our guidance.

Our next question comes from the line of Mark Zgutowicz from Benchmark.

A couple for me. Steve, just a follow-on to the question on the scaled advertiser metrics. So your scaled advertiser growth was up year-over-year, but down sequentially. And I'm just curious if that was sort of in line with your internal expectations and what sort of the yin-yang is, I guess, balanced between those 2 metrics, meaning do you expect to see more of a lagging effect on the revenue side? And could that inflect at some point this year relative to that growth that you've been seeing on the actual advertisers?

And then a second separate question, just appreciate if you could unpack your 1Q ex-TAC margin guidance. 1Q has guided 100 bps of expansion year-over-year at the midpoint. And considering that you're lapping Yahoo tests, I think that had a positive effect on margin. Just curious if you're seeing a mix shift towards higher take rate publishers or if that's being driven by yield improvements? I'll just stop there and maybe a quick follow-on to that.

So I think in terms of the scaled advertiser trends, so we tend to look at that year-over-year because there is some seasonality to that. So looking at it sequentially quarter-over-quarter, it can be pretty deceptive similar to our revenue itself, like if you look at it sequentially quarter-over-quarter, you can see some things that may look weird. But if you look at it year-over-year, a lot of that normalizes. So I tend to look at it year-over-year. I will also just say that there is some -- the metrics bounce around a bit in any given quarter. So they're tough to predict on a quarterly basis.

So for instance, if some of our bigger advertisers get really aggressive one quarter, they can squeeze out some smaller advertisers just because they're willing to bid more. So they're hard to predict on the numbers basis. But I think if you look at it year-over-year and over a longer period of time, then I think it tends to normalize. So that's the way we tend to look at it. We tend not to look at it quarter-over-quarter sequentially as much.

In terms of our revenue ex-TAC guide, I think the simple answer again to your question about like are we seeing kind of just traction in higher-margin areas? The answer is yes. So I think we're seeing kind of a shift in our business to higher-margin areas. It's -- Connexity, for instance, is 100% ex-TAC. So if business shifts to them, that appears as higher ex-TAC business to us, ex-TAC margin business. But also, to your point, it's also just in between regions and specific publishers, the mix is just trending in a positive ex-TAC margin direction. So it's less to do with increasing yields right now, although I'm hopeful that we'll see that also over time. It's more mix of business today.

Okay. Got it. Appreciate that. And if I could just ask maybe one more, just maybe zeroing out here a bit. If you look at your rest -- and this is more a topic on just geo expansion, generally speaking, but your rest of world is roughly 35% of revenue, and that grew quite nicely in 4Q. It was up about 10%, which looks like it's the fastest growth you've seen in fourth quarter ex Germany. I'm just curious if you can maybe talk about any dynamics at play there in '26 that -- and how they compare to '25 in rest of world.

Yes. I mean we're seeing nice -- and by the way, you asked about margin and mix of business. That's part of it. Some of those other geos tend to be high margin for us. So as they grow, they tend to help with our overall margin picture, ex-TAC margin picture. So that's part of it. But also to your question about anything that we're seeing there. So we're seeing nice growth internationally. If you remember, we used to be about 40% U.S., 60% rest of world. And once we brought on Yahoo, we got back closer to 50% U.S., 50% rest of world. I think this past quarter, it was 47% U.S., 53% rest of world.

I think we're going to continue to see faster growth internationally than we will in the U.S. And that's just kind of normal because a lot of those markets we're still newer in. So we have more growth opportunities in a lot of those markets. So I think that's going to continue to be true as we go forward. So I think what you're seeing there is basically just a dynamic of less mature markets versus more mature markets and higher growth in the less mature markets.

Our next question comes from the line of Zach Cummins from B. Riley Securities.

So just two for me. The first one, I thought it was a notable call out that your, I guess, we'll say, non-scaled advertisers still contributed about 1% to growth here in Q4, largely due to early adoption of the Realize platform. So any incremental data you can give around kind of how you're ramping the testing process, what tends to work best when quickly scaling up from these tests to expanding to more full budgets for some of these advertisers?

And then second question, Steve, it seems like we have a greater shift of adjusted EBITDA going into the second half of this year versus what we saw in 2025. So can you give some context around maybe timing of investments or other factors we should consider when modeling that out?

Sure. So I think on your first question about the non-scaled advertisers, it was an interesting effect. So we saw a lot of testing budgets in Q4, and that is -- that drove 1% incremental growth, which is the first time you've seen that. In fact, if you look at the full year, non-scaled advertisers were basically down a bit year-over-year. So Q4 was unusual in that regard. But I think it's encouraging because at the end of the day, what we do want is a bunch of advertisers coming on to test our platform. So -- and Q4 is a good time for a lot of them to do that because it's where they have some of their maximum budgets and they're looking to test new things. So we found it encouraging. I'm hopeful that, that translates into more revenue going forward, although we're not counting on that, but it was encouraging to see that. So that's kind of what we saw there.

In terms of the EBITDA question that you had, I think the biggest impact on our EBITDA in Q1, in particular, is that it is -- we have a headwind from foreign exchange rates. So I think I mentioned that in my prepared remarks that the -- we have about an $11 million headwind on OpEx as we head into 2026 due to foreign exchange rate, mostly the Israeli shekel, that hits first quarter and second quarter much more heavily than third quarter and fourth quarter because of the fact that if you look at how the shekel declined over the course of 2025, it really took a nosedive starting sometime in Q3. So that's one factor.

We're also intentionally up putting some of our marketing expense, especially where we're marketing to advertisers upfront in Q1 and Q2. So that's part of it. But in general, I think, obviously, our guidance reflects what we expect to happen over the course of the rest of the year on OpEx. And we do have some efficiency initiatives that are going on that we think can help us in the second half. So I think it's a little bit of some upfront costs that we knew were going to happen, foreign exchange rates and then us expecting to get more efficient as we go through the year.

Our next question comes from James Kopelman from TD Cowen.

First one for Adam. Given some ongoing macro uncertainty and state of the U.S. consumer, what's your sense of conditions in the overall digital ad market? And what are you hearing from your conversations with advertisers regarding their plans for budget growth this year?

And then another one for Adam. I just want to ask about the ARPU opportunity for publishers adopting AI on their sites. Where are we in that process? And what kind of progress are you seeing with publishers so far? And then I'll follow up with Steve as well.

So in general, I think there's a significant kind of trend in the industry at large towards performance advertising. I mean if you saw last year, we announced 2 extended partnerships, one with Paramount and one with LG. These are TV, big TV broadcasting companies that we're honored to be working with. And those partnerships are primarily around more ways for TV advertisers to get mid- to low funnel metrics by working with Realize. That's a whole new type of demand opportunity for us. And remember, TV is a $100 billion market just in the U.S. So if we can take a piece of that and prove much like I think Amazon is doing such a good job with Prime, showing that you can buy TV and at the same time, through amazon.com and the rest of their consumer journey, you can show that TV drive mid- to low funnel metrics.

And the reason I'm saying that is because I think for Taboola, there is a lot of growth opportunity because when you go beyond search and social, I think we can truly become kind of the monetization layer for the open web, the company that any advertiser and any company that's not Google and Facebook, who needs someone that can generate conversions to work with. And Taboola is, I think, to my knowledge, the biggest and best conversion machine outside of Google and Facebook. So I think there's going to be a lot of growth for us in different ways, working with different types of companies as demand source. And it's really nice to see companies like Paramount and LG, and you'll see more throughout the year kind of partnering with us and spending more with Realize.

So I think we're on the right side of the industry. I think it's going to be much, much harder to be in the full funnel space or specifically in the top of the funnel space. And it's going to be much, much more important and critical, especially in this world with tariffs and things to be the go-to company for anything outcomes, anything measurement, anything performance. So that's about that.

And the second thing about ARPU, which I had such many great conversations even yesterday with some of our bigger publishers. What we're seeing is that -- so two things. One, when consumers ask questions on a publisher site, they essentially become super -- if you become a super human, a super engaged consumer, you're much more likely to engage with an ad, you're much more likely to engage with a piece of content. You're the best version of yourself. And that's probably why I believe Google is very excited about Gemini because if Google sees what we see with DeeperDive on publisher sites, they know what we know, which is it's a very lucrative piece of interaction with consumers for advertisers.

And the second thing we're seeing, and we'll share more data about that later in the year, is that when advertisers show up in LLM experience, the opportunity for them to drive conversion and the CPMs we're seeing are something that I can tell you in 15 years of doing this, I have never seen before. So if we can scale that, if we can create a habit for consumers to talk to publishers they love about -- I mentioned the Knicks, I love the Knicks, I will never spend 5, 10 minutes watching highlights and talk to ChatGPT about the Knicks, never going to happen. But I do this every morning with my kids. We watch ESPN. We get the highlights. We read about it. It's something we like to do.

So if we can get those trusted loved publishers to offer AI so consumers can talk to them, I think that's going to be a beautiful future for them and for us and for advertisers. So the question is, can we create that habit? And it's early stages for us, but I'm encouraged by what I'm seeing.

Great. And then just quickly for Steve, reduced the share count pretty significantly over 2025. Going forward, how are you thinking about balancing investment with returns to shareholders, especially given healthy free cash flow generation? Would you expect to continue to significantly shrink the share count? And also on Connexity because you threw that in there a couple of questions ago, I just want to follow up. Any color on e-commerce growth, how that's trending relative to the rest of the business?

So in terms of capital allocation and share buybacks, so we continue to expect to use the majority of our free cash flow for share repurchases. So I think we've said that we expect to convert 60% to 70% of our EBITDA into free cash flow. So we -- and then we expect to use a majority of that to buy back shares. So if you kind of look at our numbers and what we're guiding to this year and you do the math, you can figure out how much we're expecting to buy back roughly. We have $180 million left in our authorization. So we've got plenty of capacity there, and that's where we expect to use most of our capital. I will note, and we've talked about this in the past that there's a chance that we may do small M&A. It wouldn't be large, but it would be something that's more of a tuck-in acquisition. But beyond that, we expect to use the majority of our free cash flow for share repurchases.

To your second question about e-commerce and how that's doing in Connexity relative to the rest of the business. So generally speaking, it's growing in line with the rest of the business. So I think we're -- they had a big Q4 for us, which was great. And I think generally, we expect them to grow in line with the rest of our business. It's our biggest ICP segment, ideal customer profile segment is e-commerce. So that's also great. Like it's our -- it's where we see the most success out of any of our ICPs right now. So it's generally -- it's a strong performing part of our business.

This does now conclude the Q&A portion of this session. I would now like to turn it back to Adam Singolda, CEO, for closing remarks.

Thanks, everyone, for being with us this morning. As you can tell from our excitement, 2025 was not just about beating the numbers, it was a turning point for the company. It's a clear validation that Realize is working on our way to become the monetization layer for the open web. As Realize continues to gain traction with our proprietary intent data and deep distribution across the open web, all of those things make us really special and it makes us different in an AI-driven world.

We believe these structural advantages position us to build and win the opportunity to become the leading performance advertising company beyond search and social. We're still early, but we're operating with a lot more clarity and more urgency than ever. Our focus remains simple: make new advertisers stay and get existing ones to spend more. Thank you all for the trust and partnership, and we look forward to spending time over the next few weeks.

Thank you for your participation in today's conference. This does conclude the program. You may now disconnect.

"

"

"

"

" TD Cowen, Research Division

" Oppenheimer & Co. Inc., Research Division

" Needham & Company, LLC, Research Division

" The Benchmark Company, LLC, Research Division

" Citizens JMP Securities, LLC, Research Division

" B. Riley Securities, Inc., Research Division

"

Good day, and thank you for standing by. Welcome to Taboola's Third Quarter 2025 Earnings Conference Call. [Operator Instructions] Please be advised that today's conference is being recorded. I would now like to hand the conference over to your first speaker today, Jessica Kourakos, Head of Investors.

Thank you, and good morning, everyone, and welcome to Taboola's Third Quarter 2025 Earnings Conference Call. I'm here with Adam Singolda, Taboola's Founder and CEO; and Steve Walker, Taboola's CFO. The company issued earnings materials today before the market, and they are available in the Investors section of Taboola's website.

Now I'll quickly cover the safe harbor. Certain statements today, including our expectations for future periods, are forward-looking statements. They are not facts and are subject to material risks and uncertainties described in our SEC filings. These statements are based on currently available information, and we undertake no duty to update them, except as required by law. Today's discussion is also subject to the forward-looking statement limitations in the earnings press release. Future events could differ materially and adversely from those anticipated. During this call, we will use terms defined in the earnings release and refer to non-GAAP financial measures. For definitions and reconciliations to GAAP, please refer to the non-GAAP tables in the earnings release posted on our website. With that, I'll turn the call over to Adam.

Thanks, Jessica. Good morning, everyone, and thank you all for joining us today. We're pleased to report another strong quarter, our third consecutive quarter in 2025, exceeding the high end of our guidance. Our new performance platform Realize is beginning to work for both advertisers and publishers. We're seeing an inflection point in our business and have greater confidence than we did even 90 days ago that we will get back to double-digit growth over time. This momentum gives us the confidence to once again raise our full year outlook. We've bought back 14% of the company year-to-date, and we're continuing to buy back shares aggressively.

As a reminder, Taboola is one of the largest performance advertising platforms outside of search and social. Our platform Realize helps businesses get leads and grow sales. It operates similar to Google Ads or Meta ads, offering a simple-to-use platform powered by AI. The key difference is that while Google reaches users in search and Meta in social, Realize engages 60 million people every day across the open web on partners like Yahoo, NBC, ESPN, USA Today, Apple News, Samsung and Xiaomi, driving those people to action. Our competitive advantage lies in our AI and first-party data drawn from what people actually read about versus what people idealize themselves on social media, giving advertisers authentic insights into users' intent and high-performing outcomes.

In 2025, we expect nearly $2 billion in gross revenue and more than $700 million in ex-TAC gross profit, which is what we keep after we pay our publisher partners who show our ads to their users. We expect to generate over $200 million in adjusted EBITDA at a 30% margin with strong free cash flow. As advertisers see diminishing returns on search and social, they look for scalable performance-driven alternatives like Realize. Taboola is uniquely positioned to take share in what we estimate is a $55 billion opportunity. Now let's turn to our Q3 results, which came in ahead of the high end of our guidance across the board.

We delivered revenue of $497 million, ex-TAC gross profit of $177 million and adjusted EBITDA of $48 million, representing a strong EBITDA margin of over 27%. We also generated $46 million in free cash flow this quarter and $117 million year-to-date, which amounts to a 96% conversion of our adjusted EBITDA in Q3. This strong cash generation allowed us to repurchase approximately 10 million shares during the quarter for a total consideration of $34.4 million. Year-to-date, we bought back approximately $184 million worth of shares, representing 14% of the company. Driving ex-TAC gross profit growth is our North Star. It indicates that we're providing increased value to our customers and fuel our profitability and cash generation.

In the third quarter, ex-TAC gross profit grew 6% year-over-year. The vast majority of our revenue is driven by scaled advertisers, those who spend $100,000 or more annually. As such, we guide investors to track 2 main metrics that affect our ex-TACs gross profit growth. The first is growing the number of scaled advertisers. The second is increasing average revenue per scaled advertiser. In the third quarter, we grew the number of scaled advertisers by 4% to 2,064. Our average revenue per scaled advertiser grew 11%, reflecting meaningful progress in driving advertiser success with Realize.

Realize's expanded capabilities and strong performance technology are driving these improved results. One example can be found with a major online travel company that was interested in growing their cruise business. While using Realize's advanced targeting and bidding technologies, they were able to achieve 67% lower CPCs versus Meta while driving a 48% increase in traffic to their site. This performance was so strong, this travel company increased their initial investment 10x and has now become a scaled advertiser on our platform.

Last quarter, we shared more about how our supply is differentiated. Overall, our exposure to search traffic globally remained in the single digits. And even as search traffic across the web declines, our total company traffic in Q3 grew year-over-year. This growth was fueled by strong double-digit increases in app traffic, now accounting for roughly 1/3 of our global supply, along with successful new publisher onboarding. 

We continue to monitor our traffic patterns, but at this time, it is a relatively small level of exposure. In summary, we're very happy with how the year is progressing. We think Realize can make us the leading performance advertising platform outside of Google, Meta, and Amazon across mobile, desktop, OEMs, messaging apps, and more. It is a big ambition, and the numbers make it clear that we're not there yet. 

That said, we see an inflection point in the business with Realize. And if you know his team as well as I do, you'd know we're motivated by big challenges. It is probably one of the reasons we were voted one of Fortune's Best Places to Work. We're taking on one of the toughest competitive landscape in the world in an enormous addressable market. We're hard at work, and we bought approximately $184 million worth of shares as we see the opportunity ahead of us. 

Before I hand it over to Steve, on a personal note, I want to say that over the past months, our teams and partners in Israel have shown incredible strength, resilience, and unity. Seeing things begin to come and people returning home safely fills me with gratitude and hope. With that, I'll hand it over to Steve.

Thanks, Adam, and good morning, everyone. As Adam mentioned, we've had a strong year so far. In the third quarter, we continued that momentum, delivering results that exceeded the high end of our guidance across all metrics. In the third quarter, revenues reached $496.8 million, up 15% year-over-year. We believe this growth reflects an inflection point and realizes traction in the market. I have spoken about the fact that we have a large amount of very high-quality supply, so what we need to grow our business going forward is primarily to earn new advertiser budgets. 

We started to see traction in that area during Q3 as our new ad platform is helping advertisers succeed and helping us win additional budgets. This showed up in our scaled advertiser metrics, evidenced by a 4.4% increase in the number of scaled advertisers and a 10.9% increase in average revenue per scaled advertiser. Both of which primarily benefited from realized improving retention and growing ad spending levels with existing advertisers when compared to the same period last year. 

As I have said in prior quarters, we are particularly pleased to see the number of scaled advertisers growing as they tend to be the fuel for future growth. I should note that the growth in average revenue per scaled advertiser also benefited from an easier comparison with Q3 2024 because during that period, we were testing ad formats with Yahoo, and revenue from that test was recognized as an offset to traffic acquisition costs rather than as revenue. Normalizing for that one-time test, growth in this metric was more in the mid- to high single digits range, and taken together with our growth in scaled advertisers, positively contributed to our revenue and ex-TAC performance. 

Ex-TAC gross profit for the third quarter came in at $176.8 million, up 6.3% year-over-year, including a 55-basis point tailwind from foreign exchange rates. Ex-TAC gross profit growth was primarily driven by strong growth in advertising spend, thanks to the success we are seeing with Realize, and includes strong performance from Taboola News and Bided Supply. Ex-TAC gross profit margins were down year-over-year, primarily due to the one-time testing we were doing with Yahoo last year. Notwithstanding, overall Ex-TAC gross profit dollars grew year-over-year. And as I have said previously, I focus more on growth of Ex-TAC gross profit dollars rather than the margin percentage. Gross profit for the quarter was $139 million, primarily benefiting from strong ex-TAC gross profit growth. 

As mentioned in prior quarters, gross profit also benefited from reductions in our other cost of revenues driven by lower server and network infrastructure costs, some of which came from a reduction in depreciation expenses related to our servers due to a reassessment of their useful lives. Our net income was $5.2 million, with non-GAAP net income coming in at $34.3 million. Adjusted EBITDA for the quarter was $48.2 million, reflecting an adjusted EBITDA margin of 27.3%. We continue to focus on cost discipline across the business while strategically investing in areas that support growth. This quarter, we had a $2 million headwind for foreign exchange rates versus Q3 2024, $3 million higher operating expenses, partially offset by approximately $1 million in Ex-TAC tailwinds. 

The impact on operating expenses was primarily from the Israeli shekel, where we have a large employee and expense base. Without this headwind, our adjusted EBITDA margin would have been roughly the same as Q3 2024. We also had higher-than-planned hosting costs related to certain growth initiatives, and we decided this quarter to further increase our marketing spend for realize based on the traction we are seeing.

In terms of cash generation, we had $53.2 million in operating cash flow in the third quarter and free cash flow of $46.3 million, representing 96% conversion from adjusted EBITDA in the quarter. Our free cash flow benefited significantly from a couple of factors, primarily high adjusted EBITDA margins and strong management of our working capital. Our free cash flow conversion from adjusted EBITDA continues to be over 70% over the last 4 and the last 8 quarters. Given our experience over the last couple of years, we think it is safe for investors to assume that we will convert free cash flow at a 60% to 70% rate over the longer term, which is above our prior 50% to 60% target conversion of free cash flow from adjusted EBITDA. For the full year 2025, I expect to do even better than the high end of that range.

Turning to the balance sheet. We remain in a strong financial position. We ended the third quarter with a net cash balance of $41.5 million. Cash and cash equivalents totaled $115.5 million, which more than offset our long-term debt of $74 million. As a reminder, earlier this year, we secured a new $270 million revolving credit facility, allowing us to fully repay our previous long-term debt loan while maintaining approximately $196 million in available capacity as of September 30. This facility also allowed us to reduce our interest expense by $1.6 million in the third quarter. With this facility, we can operate with a lower cash balance while preserving access to significant liquidity. We continue to believe share repurchases are one of the most compelling uses of capital.

In the third quarter, we repurchased approximately 10 million shares at an average price of $3.43 for a total consideration of $34.4 million. Year-to-date, we have bought back nearly 14% of our outstanding shares, reducing our total share count from approximately 337 million at the end of 2024 to about 291 million at the end of [Audio gap] Q3 2025. As an update to our share repurchases from Yahoo, we are no longer required to purchase shares from Yahoo for the remainder of 2025 due to meeting certain Israeli regulatory conditions. This means we have the ability to buy more shares in the open market.

Moving to guidance. For the fourth quarter 2025, we expect revenues to be between $532 million and $542 million, gross profit to be between $166 million and $171 million, ex-TAC gross profit to be $204 million to $210 million, adjusted EBITDA to range from $83 million to $85 million and non-GAAP net income to be $52 million to $56 million.

For the full year, we are raising our guidance across the board. We now expect revenues to be between $1.91 billion and $1.93 billion, gross profit to be between $550 million and $564 million, ex-TAC gross profit to be $700 million to $710 million adjusted EBITDA to be $209 million to $214 million and non-GAAP net income to be $139 million to $144 million.

This guidance reflects continued momentum across our business. I would note that in Q4, the adjusted EBITDA guidance reflects a forecasted headwind from foreign exchange rates of over $5 million on operating expenses, partially offset by ex-TAC tailwinds, which reduces our adjusted EBITDA by approximately $1.5 million and reduces the adjusted EBITDA margin by over 140 basis points.

Also, as a reminder, when you are comparing each of the quarters this year to the same quarter last year, you must keep in mind the onboarding of Yahoo, which impacts quarterly comparisons this year. As a result, we believe the full year projected growth rate of 6% at the midpoint of our new range normalizes for these dynamics and is the best representation of the true growth of our core business in 2025.

In summary, we're very pleased with our Q3 performance and the strong momentum we've built so far this year. We're seeing an inflection point with Realize and remain focused on delivering against the goals we set at the beginning of the year. There's still work ahead, but we believe we're on the right path toward achieving double-digit growth over time. With that, let's move to Q&A. Operator, can you please open the lines for questions?

Yes, Thank you. [Operator Instructions] Your first question comes from the line of Zachary Cummins with B. Riley Securities. Your line is now open.

Hi, Good morning. Thanks for taking my question and congrats on the strong results here in Q3. So just starting off with Realize platform, nice to see the incremental traction that we're seeing on that front. Just curious, in terms of taking it to the next level of capturing more of these advertiser budgets, is it more just proving it out and testing it in the market? Or what are kind of the next steps in terms of taking this from strong traction to meaningful contribution to the overall P&L?

Yes. So I think overall, we're seeing good momentum, which is encouraging for us to see. If you look at what we said when we launched it, taking a step back, there are really 3 things that move the needle financially. The first one is just going from native advertising to performance, and that takes time in order to shift kind of perception in the market and getting more advertisers to be aware of Taboola as a place, they can spend money beyond search and social. So that's one. The second thing is our focus on the sell side on ICPs. And that's important because when we sell to the right clients, we tend to see higher retention and more spend. And the third one is keep iterating on the tech front. So predictive audiences, new formats, new placements. So those 3 things, keep iterating on those, we believe will make a positive impact.

Understood. And my one follow-up question is just building on Realize -- interesting partnership that you announced with Paramount. So can you talk about the performance multiplier product and how maybe Taboola could have more of a presence in CTV over time?

Yes. So I mean, let me just say that it's financially small as of now, but very exciting. So what we've announced, which is essentially a demand generation opportunity for Taboola through Realize, we're tracking essentially the industry overall going more and more into outcome and measurement and performance-driven market. And even in television, which is about $100 billion market in the U.S., we're seeing these dynamics.

So what we've announced, which is essentially a demand generation opportunity for Taboola through Realize, we're tracking essentially the industry overall going more and more into outcome and measurement and performance-driven market. And even in television, which is about $100 billion market in the U.S., we're seeing these dynamics. Advertisers are expecting to not only enjoy the benefit of a large screen when they get consumers exposed to ads, but also track and be able to drive conversions through that campaign. 

We're seeing this with Amazon, obviously, telling advertisers if you buy Prime, we can show ads also on Amazon.com and show you that someone ended up buying the product. It's a very powerful pitch to advertisers. So what we've announced with Paramount, which is a great partner of ours on the CBSI front for a very long time and now on the CTV front, is to bring the combined power of TV and performance advertising into one home, if you will. So if you're buying an ad and that ad is shown on Yellowstone as an example, and you see that ad in your living room, you, a consumer, and people like you may be through a matching integration with Paramount, may be seeing relevant ads on the open web through Realize. 

And the point here is that we're able to expand the audience. So even if one person saw an ad on TV, we're able to show maybe 10 other people ads that are relevant to the same advertiser and then report back, and this is the exciting part, reports back to the advertiser, how many clicks have occurred, impressions, conversions, price per acquisition. So, you're buying a TV and you feel like you're buying Meta. I'm excited about it, and I hope we'll continue to see traction, but it's like retail media for TV, if you will, it's still early days.

Your next question comes from the line of Laura Martin with Needham.

Great numbers, guys. I have a couple. Can you talk about what's going on with traffic? And if you're not seeing any degradation in traffic, I'm very interested in the quality because these LLMs are searching sort of anywhere between 8,000 and 8,000 pages per query to give an answer. So my first question is on traffic, both volume and quality of is human traffic declining even if total traffic to your site is not? And then my second question is, you guys were really early adopters of using AI to improve your yields and conversion rates. I'm interested in an update on actual metrics of where you've seen improvements either in costs or in revenue or in yield from your updated AI implementations, please?

Sure. Thanks for the question. So I can start, and we can bounce off of this. So the first thing we're seeing actually surge in traffic overall, which is interesting, obviously, as you say, in times of search traffic declining because of L&M engines. The main 2 reasons we're seeing traffic going up is, I would say, one, we're seeing an increase in direct traffic, actually specifically through apps. So many of our partners have a very strong brand name. So if you think about the ESPNs of the world, the CNBCs, Yahoo!, and OEM partners such as Apple and Samsung, and others, many of them have a very strong brand and are recognizable with consumers. 

And while search traffic is going down, we're seeing their app direct traffic going up, and that affects our overall traffic mix. So that's something that's really helpful. And the second thing is, we're doing a good job onboarding new partners. So think of new publishers and new devices through Taboola News, we've seen positive trends in adding new partners. So for those 2 reasons, overall company-wise, we are seeing growth in traffic.  So, that's about that. 

Human traffic, we're not seeing any material change that I'm aware of. In general, the way we're doing it today, we have this index that actually allows us to constantly make sure that our supply has high quality for advertisers. When we onboard a new partner or as things change, we're able to see if the conversion rate from a certain publisher or the conversion rate from a certain page has changed. And if it's within a certain range of what we expect, we call that human and good. And if it's too far down, we actually part ways with that partner. We're not sustaining partners on our network that do not perform to advertisers. So we've been doing this for a long time. And I think advertisers really appreciate that because they know they can always rely on our traffic being high quality. Steve?

Yes. And then in terms of your question about AI, and where are we seeing impact from that? So just to take a step back real quickly, we use AI across our business. We've been a deep learning-based business for better than a decade now. And that is what the technology basically drives, how we decide which ads or which pieces of content to show to which users in which particular context at any given time. So that is a key part of our business and has been for a decade. 

LLM-based AIs, where we started using, obviously, more recently, as they become more advanced, we use those primarily in 2 areas of our business. So one is we announced, obviously, Laura, you remember we demoed Abby with you, where we showed an LLM-based assistant for advertisers that helps them to get up and running on our platform. And now, over time, it is even helping them to optimize campaigns and to do more with the platform. So we use it there. 

And then internally, we're using LLM-based technology a lot more for productivity reasons. So a great demo I saw the other day was showing our sales team where they can say, "Hey, I'm about to go into a meeting with Nestle with this person, and an LLM-based tool will actually prep a whole kind of package of information that they should know. What has Nestle been saying about their goals as a business? What does this person do?” It's kind of amazing how much information they get going into the meetings that they come in prepped. So we're using it as a productivity tool in that way. 

In general, what I would say is that we will always see the biggest impact on our business with AI is anything that drives higher yields and higher success rates for our advertisers. And that's true. So when we talk about seeing an inflection with Realize, frankly, a lot of that is coming on the back of better algo, things like predictive audience is an AI-based prediction tool of where could you get more conversions on our network and what do you have to do to get those. So that's probably impact number one. Abby is definitely having an impact on our advertisers. So we're seeing more success, thanks to that. And then I'd say that we are seeing productivity, but we're probably earliest in that area in terms of where we're seeing impact.

And lastly, Laura, just to add one more note is that we also launched deeper dive a few months back, which is our kind of ChatGPT for the open web driven by advertising as a revenue source. Still early days, but that's another thing that can generate surge in quality traffic, definitely human traffic because people have to type and engage with it. So that's another investment we're making to try to create more quality supply for advertisers in the form of LLM and support publishers in growing and getting into the AI era as we know it.

Your next question comes from the line of Jason Helfstein with Oppenheimer.

I was on a few calls, so I apologize if this was already covered. But I'm struggling to understand kind of, again, why revenue ex-TAC on a year-over-year basis will decelerate, call it, seven points? And then why the other cost of goods, the non-TAC COGS is going to be up $5 million sequentially? So can you unpack that? And then I've just got a question about next year.

Yes. When you talked about the non-TAC COGS, are you talking about the other cost of revenue?

Correct. Yes.

Okay. Yes. So I guess I'll start with kind of the general discussion of ex-TAC and where we are with ex-TAC. So first of all, I think margins were down -- when you look at ex-TAC margin, they were down year-over-year, primarily due to the Yahoo testing last year. So remember, last year, we had a format testing with Yahoo where it was recognized on a net basis. Basically, it was a TAC offset. And that distorted our margins year-over-year. So I think if you're talking about a margin in Q3, it was down because of that.

If you're looking at ex-TAC dollars, they were up 6% year-over-year in Q3, and that's kind of what we focus on more as the dollars. Q4 is going to be down year-over-year. That's mostly due to two factors. One is there was a -- the Yahoo onboarding last year basically caused some distortion between quarters. So that's part of it. And then we also had particularly strong demand from Chinese advertisers in Q4 last year that, frankly, was unusual. And with the tariffs this year, they did drop. We talked about that in Q1, and they have not gotten back to the same rates yet. So those are the primary two factors for Q4.

In terms of the other cost of revenue and changes there, so first of all, we've changed the accounting on our servers. That actually has brought down other cost of revenue. But we have also written off or not written off, we've changed the way we're accounting for some of our capitalized projects. I believe that's a big part of the remainder of it.

Okay. And then just philosophically, I think we were previously thinking next year would grow low single digit. I mean, obviously, coming off this -- again, for revenue ex-TAC with this minus kind of 2% guide for the fourth quarter, is it still fair to assume that you think the company could generate positive growth in revenue ex-TAC for next year?

Yes. So yes. And I think that generally, the way I think about it is, first of all, we're seeing good momentum with the business. Like I think we're happy with what we're seeing. I think I wouldn't look at quarters when you're thinking about the growth rate. And while we'll give you guidance, obviously, for 2026 in February, what I would look at now is our full year growth rate for 2025 is a good proxy for, I think, where you can start with thinking about 2026.

Your next question comes from the line of Mark Zgutowicz with The Benchmark Company.

Steve, you talked a little bit about marketing spend. I'm just curious, if you think about Realize marketing spend returns, what adjustments have you made since launching Realize? Maybe you can talk about what's worked, what hasn't and perhaps how your sales capacity is there relative to where you'd like it or if it's where you are comfortable with? And then in terms of margins, if we look at the fourth quarter adjusted EBITDA margin guidance, is that a good proxy for us to think about in terms of 2026, excluding any potential rev ex-tech acceleration? And then maybe a last one for Adam, if I could. If you think about opportunities for investment alongside Realize incrementality, can you maybe prioritize Agentic, MCP, ad CP or any other areas?

Sure. Thanks, Mark. So I'll answer the first two. So first of all, in terms of the marketing spend, yes, so I mentioned, obviously, in my prepared remarks that we've increased marketing spend intentionally over the last two quarters, Q3 and now heading into Q4. The reason we did that is we're seeing and again, it has to do with the inflection we're seeing with Realize. We're seeing that advertisers are more likely to be successful coming on to our network now with the launch of Realize and a lot of the new product features that are baked into Realize, they're more likely to be successful.

So obviously, what that means is if I previously was spending and I'm making up numbers here, so don't take these -- don't build these into a model. But if I was previously spending $1,000 to get an advertiser onboarded in the U.S. through -- in terms of marketing spend, and they were -- had an X percent likelihood of being successful and then eventually having a certain lifetime value, that X percent likelihood of success has gone up, which means the lifetime value is still similar, maybe even improved a bit, means I can spend more on marketing, and I can basically still have a good positive ROI on it. So as we've seen that, we've started increasing our marketing spend, which has a smaller effect on the immediate term, but again, brings on more advertisers, which should help us grow faster in the future.

So that's kind of what's going on with the marketing spend. In terms of specific areas that we're seeing that, geographically, it's been pretty broadly diverse, so globally, although we are seeing particular momentum in the U.S. in that regard. And then in terms of the types of advertisers, it's really those -- we talked about ICPs or ideal customer profiles where we talked about the launch of Realize. It's those verticals. It's finance, it's auto, its health, it's direct-to-consumer products. It's those verticals that we're seeing the best traction.

In terms of your second question about expectations for adjusted EBITDA and kind of where -- what you should think in terms of that going forward, I think -- so Q4 is a seasonally strong quarter from an adjusted EBITDA margin perspective. So don't look at Q4. But I think what we've said repeatedly in the past, and I think it's still true today, we use 30% adjusted EBITDA margins as kind of a guardrail for ourselves to think about what we should -- how much we should invest in growth versus how much profitability we want. And I think that's a good starting point for how to think about our margins and our OpEx for next year.

I can take that AI one. So, Juan, on the first part of your question, I think all of what we do is investing in AI in terms of driving primarily advertiser success. Realize is drive most of our revenue as a company now and making advertisers successful through Realize will be the main way we will go back to double-digit growth as a company. So -- and then Realize is -- we're tracking Realize, as you know, with scaled advertisers, how many of them do we have and then how much they're spending in average. The reason that's important is because if you do a good job making advertisers successful with unique data, a lot of distribution and advanced AI investment, then you should see more scaled advertisers and you should see them able to spend more because it means you have higher retention rates. People try you, they churn less, which means they get retained. And then once they're in, they're spending more and more over time.

That's what you want to see. And as you saw in my remarks, we're seeing a 4% increase in scaled advertisers and 11% growth in average spend. So I think for us, it's laser focused on how do you get return on ad spend as fast as you can for advertisers so they don't churn. And then what technologies such as predictive audiences and other new formats and new supply, you can provide them as part of our technology so they can spend more and more over time. And that's -- these are the 2 areas of focus for us. When I think about Taboola and its position in the marketplace, you asked about MCP, I think we're well positioned because the vast majority of our revenue comes from advertisers who buy from us directly.

That means that we're able to share more with publishers and keep a very healthy margin for us as a business. And that's not the traditional ad tech usually diagram where you have companies that are doing one side of the marketplace, and a lot of money gets lost along the way. So I think for us, it's important to track MCP in this industry and companies such as Taboola, which is a 2-sided marketplace. Publishers who work with us, work with us directly. Advertisers who buy from us, buy from us directly. This is similar to almost a consumer company. Only for us, we don't have our own Instagram. We have reach to consumers through publisher relationship. So I think we're well positioned, and you can see that in our performance.

Your next question comes from the line of James Kopelman with TD Cowen.

The first one is for Adam, just following up on Laura's question on traffic. You mentioned that app traffic is now 1/3 of supply. Where do you see that trending over time? Do you think that could hit 50% or higher of supply? And do you expect double-digit app traffic increases in the fourth quarter as well? And then I have a follow-up question for Steve.

Yes. I mean it's encouraging to see that traffic being already 1/3, almost 1/3 and because that's a good stable base that is not affected by search as much or at all. And so I think that's already encouraging from our perspective as I look into the future. I think it has a chance of going up faster because, one, it's a much more engaged audience tends to be. So if you talk to publishers, they'll usually tell you, and we see it that an app user is a significantly more engaged consumer. They spend more time, they read more, they generate a lot of revenue. So publishers are always motivated to move people into app. And with -- again, with the risk of LLM, I think I'm betting we'll see more dynamics of publishers trying to get consumers to download the app and using that.

So I think that's going to be a positive trend. And two, -- we're seeing a lot of growth coming from Taboola News and in-app monetization through partners like Apple News is obviously an incredible one in Samsung. And we spoke earlier this year about LINE, which is a messaging app. So I think for us, we have aspiration as a company to keep working wherever consumers spend real time. So that over time, we have aspiration to be on every device, every lock screen, every swipe you may have, provide news if relevant as part of your utility app. So that's something that we invest in as a company. And I think there's a good product market fit with what we can provide, which is content, data and revenue to what the market wants. So I do suspect this will have a positive trend.

And then for Steve, in the third quarter, you were able to keep OpEx expense growth at a moderate level even as you accelerated growth in the business. What are your thoughts on investments and how we should think about headcount and expense growth over the next couple of quarters into 2026 as well, particularly as it relates to operating margin in the business?

Yes. No, good question, James. So I think, generally speaking, the way we've always looked at our business is that we should invest in growth, obviously, where we think we have a positive ROI, but limit ourselves by saying that we'd like to always maintain a 30% plus EBITDA margin.

So I think, generally speaking, the way we've always looked at our business is that we should invest in growth, obviously, where we think we have a positive ROI, but limit ourselves by saying that we'd like to always maintain a 30% plus EBITDA margin. So I think that's the way we'll still think about things going forward. And that's the way we'll probably plan 2026 as well. I think in terms of how to think about like what -- where do we invest and what are we going to spend our money on, I think it's really realized is going to be the primary area of investment for us going forward. Obviously, we always have, Adam, you used to call them speedboat type of initiatives, but we have things like deeper dive where we're investing a small amount of money just to see what we can learn and figure out. But the big investment is going to be realized going forward just as it has been. But I think you can expect the operating expenses to grow in line with growth and for us to maintain 30% plus type of EBITDA margins.

Your next question comes from the line of Tyler DeMatteo with BTIG.

Adam, I wanted to come back to some of your comments at the beginning of the Q&A on kind of the sales approach. What's the biggest opportunity on the sales side of things to improve the brand perception and ultimately kind of realize adoption there? Like what are some of the learnings that you've seen where you can see an incremental improvement on the sales side and that opportunity? And then my second question is on the comments about the inflection point in realize, what are some of the underlying assumptions baked into that? Is that the number of advertisers? Is that the propensity to spend, the dollar value of spend? I'm just curious like what's the underlying assumptions there?

So one, I think that's a great question. I think, one, we're investing overall from a perception perspective. We have -- we're putting our people to work in important events and things to interact with the market and put realize in the front as a way to attract performance advertising in a world that goes -- wants to go beyond search and social. So I do think that's a real need, and we have a shot at doing this in the market. So one, we're investing in being out there and telling our story.

Two, show me who you work with, and I'll tell you who you are type of thing. So if you're able through your technology and your investment to get good advertisers to be successful with you, to tell the story for you, I think that affects your story the most. Your brand is what people say about you when you're not in the room. So for me, that -- to see advertisers like the ones we mentioned working with us, excited to tell the story is a great sign. And then taking a high-level view to have -- to look at the number of scaled advertisers and the average spend going out, these are the right metrics. So all of these are good initiatives.

And like I mentioned also, our sales team now knows and salespeople go where they see money. They know that if they sell into the right segments in the market to the right advertisers tend to be those that have high consideration stage like travel, health care, auto, commerce, we tend to be financial services, we tend to be really, really good. I mean I always joke that if you run a business, a mortgage business or financial services business in America and you're not buying from Taboola, it is irresponsible. You have to try because we're very good at this, and our sellers know that, too. So by focusing on ICPs and going strong on those, we tend to see better results. So all of those initiatives are the right ones to eventually, I think, also affect the brand and the perception, but that takes time. There are no shortcuts. We spent a decade being the native advertising company, and we're going to spend the next decade, hopefully building the largest performance advertising company outside of search and social.

And in terms of your second question about like what are the underlying assumptions around the realized inflection point, I think the way to think about it is from realize, I mentioned earlier that we're starting to see advertisers have a greater likelihood of succeeding with us, greater likelihood of being able to scale with us, greater likelihood of being able to meet their goals. And I think what that means is what I want to see is that our -- I've said this in the past, I want to see our number of scaled advertisers growing year-over-year consistently. That is to the best metric that indicates that we're having success. We obviously also would like to see the average revenue per scaled advertiser grow over time. But generally, as I've said in the past, the number of scaled advertisers is what is the fuel for future growth. So that's the metric we focus on the most.

Your last question comes from the line of Matthew Condon with Citizens.

My first one, maybe just shifting gears here a little bit. Can you just talk about the Taboola News? It looked like it was another strong quarter. Just what's the sustainability of growth there? And how should we think about that contributing in 2026? And my second one is also just on your partnerships with some of the OEM partners. Just how are these progressing and scaling up here? And should we expect these also to be key contributors in '26?

So I can take the first half of that. So first of all, yes, Q3 was good for Taboola News. It's Taboola News is growing faster than the rest of the company, which is nice for a growth initiative like that. It's also -- I think the important part of Taboola News is it's part of our unique supply strategy. So it is a very unique type of supply that, by the way, obviously completely immune to LLM disintermediation and that type of issue. So that's great. It also is fresh users, good data.

So it's -- when you're on the cell phone device natively, you know a bit more about what's going on than other times. So it's all part of that getting unique supply and advertisers really like it. It's before a user gets to their social network. It's before they start browsing the Internet. It's a very unique time to meet the user. So I think it's also great for advertisers. And so with everyone wanting advertising, including cell phone manufacturers and OEMs, I think there's big upside to this. Where it goes, we obviously haven't spoken specifically about guiding to that and we don't break it out, but I think we see a lot of upside potential to this over time. Sorry, operator, are we on still?

Yes. Yes. I didn't know if there was going to be another question. So at this point, there are no further questions, and I'll turn it back to Adam Singolda for closing remarks.

Thank you. Thanks, everyone, for being with us this morning. If you take 3 things from the quarter that matter, number one, we've hit an inflection point with Realize, which is our biggest investment. Customers are giving us good feedback and our product is driving good results. It shows in our scaled advertiser numbers, a 4% increase in the amount of scaled advertisers. That's obviously a good thing. And we're seeing 11% higher average spend. And we track those 2 numbers as a proxy for realized success and realizes most of our revenue. So that is our main way to grow in the future.

Number two, we're feeling better about our financial performance. We like the direction we're heading. And as such, we bought 14% of the company year-to-date and intend to continue to buy aggressively. And number three, I'm proud of the team. We're taking upon ourselves a big challenge, and we're hard at work, and I believe we can do this. So I'm looking forward to interacting with many of you over the next few weeks, and thanks for joining us today.

Yes. Thank you for your participation in today's conference. This does conclude the program, and you may now disconnect.