Genius Sports (GENI) 2026年第三季度业绩电话会: 现金流增长, 业绩指引上调

核心要点

• Genius Sports公布2026年第三季度集团营收为2.7亿美元，高于其2.6亿美元的指引；调整后EBITDA为8700万美元，超过其8500万美元的指引。
• 调整后EBITDA利润率达到32%，高于上年同期的20%。毛利率提升至34%，上年同期为25%；GAAP营业利润为1000万美元。
• 自由现金流总计5400万美元，占调整后EBITDA的转换率为61%。若剔除1600万美元的现金利息支出，无杠杆自由现金流转换率约为80%。
• 博彩业务营收同比增长33%，主要得益于滚球投注、运营商赢率改善、合同续约定价以及预测市场的商业化变现。
• 媒体业务营收增长126%，反映出收购Legend的影响、预测市场及在线体育博彩支出的增加，以及Moment Engine的普及采用。
• 管理层将全年营收指引上调至10.15亿至10.35亿美元，调整后EBITDA指引上调至2.87亿至2.97亿美元。

核心财务数据

业务与运营表现

博彩与预测市场

博彩业务营收约有70%–80%来自合同约定的最低保底收入，这为该板块奠定了稳固基础，同时保持了享受更高市场活跃度红利的潜力。

在NFL赛季的前四周，博彩总收入（GGR）实现了接近60%的同比增长。滚球GGR增长近50%，而每位用户的BetVision使用时长增加近50%。管理层表示，从赛前投注转向滚球投注依然趋势向好，因为运营商利润率更高，且BetVision等产品能够进一步加深用户参与度。

Genius Sports通过获客、实时交易数据及合约结算推进预测市场变现。该公司表示，年内在该领域新增了50多家新客户，目前已为主要交易所和机构做市商提供服务。据管理层介绍，实时交易约占体育预测市场交易量的三分之二。

该公司向做市商收取数据费及分成收入。管理层表示，自收购Legend以来，预测市场的获客成本大约翻了一番，随着各运营商竞相争夺用户，Genius Sports旗下的媒体资产因而受益。

Legend整合进展

管理层表示，Legend在商业运营、技术和企业文化层面的整合进展均快于预期。与Kalshi和Polymarket达成的协议被列为合并业务带来营收协同效应的范例。

Prediction.com将Legend的受众与技术同Genius Sports的官方数据相结合。该公司表示，使用整合了实时体育数据和交易所交易行情产品的用户，进行交易的可能性高出五倍。

Legend超过一半的营收来自此前建立的客户合作关系，其中部分合作甚至可追溯至15年前。管理层还强调了该业务在互联网博彩（iGaming）、在线体育博彩和预测市场领域的地域及产品多元化优势。

媒体与广告业务

Moment Engine在第三季度新增了279家广告主，此前第二季度新增了174家。客户包括丰田、道富银行、Whoop和达美乐比萨。第三季度净营收留存率较第二季度达到224%，不过管理层提醒称，增速可能会随广告主的采用情况及支出模式而有所波动。

媒体业务营收中，通过自助渠道产生并按净额法确认的份额日益增加。与全额法相比，这虽拉低了报告营收，但带来了更高的利润率。

Genius Sports还在通过GeniusIQ扩大以体育为驱动的广告库存。相关举措包括与克里斯·保罗（Chris Paul）合作打造由运动员主导的内容，以及通过Univision提供墨西哥足球超级联赛（Liga MX）的实时洞察与可视化数据呈现。

管理层业绩指引

管理层预计，第四季度自由现金流在绝对值上将与第三季度的5400万美元相当（受具体季度结算时间影响）。偿还债务仍是公司的首要资本配置方向，预计2027年净杠杆率将进一步走低。

Genius Sports维持其2028年目标不变，即调整后EBITDA利润率达到35%，且无杠杆自由现金流转换率至少达到60%。管理层表示，目前每年的利息支出为5000万至6000万美元，意味着在2028年目标水平下，自由现金流转换率将超过50%。

该公司预计将在2027年向无杠杆现金转换目标迈出实质性步伐，但将在下一次财报电话会议上提供详细的2027年展望。

风险与关注重点

• 预测市场监管：管理层预计监管与税收结构将持续演变。监管权可能仍归联邦层面，也可能转向各州框架，但具体时间节点和最终结构尚不确定。
• 广告业务波动性：Moment Engine的采用尚处于早期阶段，客户支出可能会因广告测试活动、季节性因素及采用模式而出现波动。
• 会计核算结构：更多使用自助媒体产品意味着更多营收按净额法而非全额法确认，尽管利润率有所改善，但这会影响报告营收的同比可比性。
• 巴西市场：管理层表示，由于合同保障条款，巴西业务的关停并未对底层业务造成实质性影响，尽管这可能在一定程度上削减了部分潜在增长空间。
• 杠杆率与利息成本：公司仍将重心放在降低负债上。第三季度现金利息支出为1600万美元，年利息支出约为5000万至6000万美元。

分析师问答亮点

• 博彩业务增长：管理层将33%的营收增幅归因于各维度的广泛强劲表现，包括合同续约定价、滚球博彩、运营商经济效益，以及来自交易所和做市商的预测市场营收。
• 2027年增长基石：公司将预测市场、互联网博彩（iGaming）、Legend、产品渗透率以及合同定价列为营收驱动因素。成本增速放缓预计将支持利润率进一步扩张。
• 数据版权成本：与NFL和英超联赛的主要协议已锁定至本年代末。管理层表示，近期新增版权的影响已消化完毕，Genius Sports无需达成额外版权交易，从而能够保持审慎筛选。
• 资本配置：短期内偿还债务优先于股票回购。管理层预计2027年和2028年更低的杠杆率将带来更多灵活性。
• NFL预测市场：公司针对2026年、2027年或2028年的业绩指引中，均未计入任何与NFL相关的预测市场业绩贡献。

财报电话会议完整实录

---

完整财报电话会议逐字稿

管理层陈述

Operator

Hello, everyone. Thank you for joining us, and welcome to Genius Sports Third Quarter 2026 Earnings Results. [Operator Instructions]

I will now hand the conference over to Genius Sports. Please go ahead.

Unknown Executive

Good morning, and thank you for joining. Before we begin, we'd like to remind you that certain statements made during this call may constitute forward-looking statements that are subject to risks that could cause our actual results to differ materially from our historical results or from our forecast. We assume no responsibility for updating forward-looking statements. Any such statements should be considered in conjunction with cautionary statements in our earnings release and risk factor discussions in our filings with the SEC, including our annual report on Form 20-F filed with the SEC on March 17, 2026.

During the call, management will also discuss certain non-GAAP measures that we believe may be useful in evaluating Genius' operating performance. These measures should not be considered in isolation or as a substitute for Genius' financial results prepared in accordance with U.S. GAAP. A reconciliation of these non-GAAP measures to the most directly comparable U.S. GAAP measures is available in our earnings press release and earnings presentation, which can be found on our website at investors.geniusports.com.

With that, I'll now turn the call over to our CEO, Mark Locke.

Mark Locke

Thank you, and good morning, everyone. This is an important and exciting quarter for Genius Sports. Against the backdrop of significant attention on prediction markets, the Legend acquisition and our cash flow potential, we're pleased to report our first full quarter as a combined business. This earnings report proves our critical position in the ecosystem, our successful ongoing integration of Legend and our ability to generate meaningful cash flow.

Our strategy is working and it's translating to financial outperformance. Group revenue of $270 million exceeded our guidance of $260 million and group adjusted EBITDA of $87 million also beat our guidance of $85 million. This resulted in a 32% adjusted EBITDA margin or 12 percentage point expansion compared to last year.

Importantly, we generated $54 million of free cash flow, representing 61% conversion from adjusted EBITDA. From here on out, you will hear me reference free cash flow as an important metric to demonstrate the earnings power of the combined business. I'm holding my team and myself accountable to free cash flow generation because we are confident that today's results set a foundation that we will build upon in the quarters and years ahead.

This performance reflects the strength of our commercial model, predictable revenue, multiple opportunities to grow across the sports ecosystem and increasing operating leverage as we scale. Today, we will cover 4 key topics: one, the rapidly growing demand for sports wagering spanning traditional online sports betting and prediction markets; two, an update on Legend, which is continuing to perform strongly; three, our continued momentum in the combined media space; and finally, four, the importance of free cash flow as a key measure of operating performance and the value that we create for shareholders.

I'll start with our Betting business, which increased 33% year-over-year. This remains a highly reliable and predictable revenue stream with approximately 70% to 80% of betting revenue coming from contractually fixed minimum revenue guarantees. Our contracts have always provided strong downside protection while preserving the upside, creating more consistent and durable revenue streams as the market evolves. That upside is driven by a diverse set of factors, each of which moved in our favor this quarter.

The NFL season started late, so this quarter only captures 3 weeks of it. But looking at our first 4 weeks of the NFL season year-over-year, gross gaming revenue is up nearly 60%. In-play betting was a particularly strong contributor as improving bookmaker win margins translated to nearly 50% growth in in-play GGR. BetVision, our watch and bet product is part of that. We are seeing record levels of engagement through our BetVision product, including with the NFL's in-market and national games on mobile.

During the first 4 weeks of the NFL season, minutes per user increased by nearly 50% year-over-year. This means fans are spending more time with the product and engaging more deeply with the live game experience. This creates more opportunities for in-play betting, an area driving significant value for Genius and its bookmaker partners.

Our bookmaker contracts, many of which we recently renewed, provide a stable base of reoccurring revenue with opportunities for both parties to benefit from future growth.

We've been intentional about building a strong foundation to support a predictable and resilient Betting business. We're now leveraging that same foundation of data and infrastructure to capture new growth opportunities, most notably in prediction markets. We believe this is the biggest opportunity in our industry right now and in financial markets more broadly.

The scale of investor interest is already evident in the private market. Kalshi and Polymarket have a combined valuation of approximately $60 billion based on their latest funding rounds. That is a striking indication of the expectations surrounding this category. For the public market, we believe Genius offers a differentiated way to invest in this growing category. The reason is simple. As dollars flow through prediction market ecosystem, we generate revenue at every step along the way.

It starts with customer acquisition. Prediction market operators are competing aggressively to acquire customers, establish their brands and gain market share. This is an area where our Media business already has proven track record, helping sportsbooks acquire customers at scale.

The combination with Legend strengthens these capabilities considerably. Legend owns and operates established products that attract highly engaged sports fans, representing some of the most valuable audiences in gaming. As operators and prediction exchanges compete for those customers, our Media properties and marketing capabilities become even more valuable.

Importantly, we support nearly every operator across traditional sports betting and prediction markets. Therefore, as customer acquisition spend increases across the industry, Genius takes a share regardless of which platform wins.

Once acquired, those customers trade sports contracts on the exchanges where liquidity is critical. Much of that liquidity is provided by institutional market makers who rely on real-time data to set the price for those contracts. This is particularly important in live trading, which we estimate accounts for roughly 2/3 of total sports volume. As more market makers enter the space and as trading activity grows, demand for our data increases. We already supply many of the major institutional market makers providing that liquidity, meaning Genius shares in the growth of the market regardless of where that liquidity comes from.

The final step is in settlement, the moment the contract pays out. Exchanges rely on official data, a single source of truth sanctioned by the league to settle contracts quickly and accurately. So as more sports contracts are traded, Genius takes a share regardless of where those trades occur.

Customer acquisition, trading, settlement, 3 critical stages of the prediction market ecosystem with Genius already monetizing all 3 with over 50 new customers this year and counting. And we're doing it using the same data, technology and infrastructure that have underpinned our sports Betting business for 25 years. That's the operating leverage of our business model.

The rise of prediction markets reinforces our original investment thesis. Genius provides data and services to the entire regulated sports wagering ecosystem and is positioned to benefit regardless of which platforms ultimately succeed. The thesis hasn't changed, only the size of the opportunity has. And I want to build on that point and clarify how we think about the size of that opportunity.

Genius serves regulated sportsbooks, major prediction market exchanges and the market makers that provide liquidity to those exchanges. Our opportunity is, therefore, linked to the total dollars legally wagered on sports regardless of the platform. Prediction markets have clearly proven the demand that exists for sports wagering in America. These platforms are simply bringing more dollars into the sports wagering ecosystem, which increases our TAM and reinforces our confidence in the potential upside to our 2028 guidance.

This brings me to the question surrounding the regulatory future of prediction markets. After 25 years in the industry, my view is that today's structure will evolve. Rules around access products and taxation will change. Some of today's advantages will shift and some of today's winners may not be tomorrow's.

We often get asked what it would mean for Genius if oversight moves from a federal to a state-by-state model. Our view is that the underlying opportunity remains meaningful either way. We're already seeing states take steps towards taxing or regulating this activity. North Carolina is taxing exchange trading fees. Illinois has proposed a transaction levy and tribes in California has signaled interest in gaming regulation. These are early indicators of how oversight could evolve.

Given the scale of the activity, we believe governments will have strong incentives to establish a regulatory and taxation framework. Whether that framework ultimately operates at the federal or state level, we expect the activity to continue in some regulated form.

The question is how it will be governed, not simply whether it will exist. So to put it simply, the ground is shifting, but we do not need it to land any particular way. If it flows back to the sportsbooks we already serve, we win. If it stays on prediction market exchanges, we're already there, too. There's no version of this that doesn't pay Genius.

Our business is built to engage fans and provide our partners with mission-critical tools to monetize that engagement. Prediction markets give us an opportunity to expand that role. This has always been the core investment thesis for Genius. What's changed is the size of the opportunity in front of us.

Now Legend. We're even more excited today by the asset that we acquired, the talented people and long-standing customer relationships that underpin its superb track record of cash-generative growth. We are now proving this through execution and results. Integration is progressing as planned across commercial, technology and culture. And as a result, we are realizing revenue synergies ahead of schedule.

Our deals with Kalshi and Polymarket are high-profile examples of what the combined business can achieve. They demonstrate how Genius and Legend create greater commercial value together than either could independently. And while these deals make the headlines, we're applying the same approach across a long tail of operators in sports betting, iCasino and prediction markets.

We're also seeing the strength of Legend's audience firsthand. Since closing the acquisition, Legend has set multiple engagement records, first during the World Cup and more recently across the NFL and college football weekends. It's a consistent pattern, and it's converting to results, particularly as competition for prediction market customers intensifies. The timing of the acquisition could not have been better for Genius to capitalize on these competitive dynamics and capture a share of the increasing advertiser spend. We're also building on this success through the launch of prediction.com, a dedicated destination for prediction markets, offering a compelling suite of tools and data.

It's another example of how we combine Legend's audience and technical capabilities with Genius Sports official data to create differentiated products. For instance, one product brings real-time data for live sporting events together with a live feed of trades that are taking place on exchanges for those events. The results are encouraging. Users who engage with this product are 5x more likely to transact. That's proof that the combination works better than either business alone.

Prediction.com also gives us deeper understanding of sports fans, the leagues and teams they follow and the products and content that resonate with them. Over time, these insights can help deliver better outcomes for advertisers and attract additional advertising spend. This represents another important area of potential synergy across the combined business. Taken together, these early results reinforce our conviction in the strategic and financial value of the acquisition.

As discussed, our Media business has a strong foundation in gaming with reoccurring revenue and contractual minimums providing meaningful visibility. We're building on that foundation by expanding the Moment Engine to a broader range of consumer brands beyond the gaming space. The opportunity is clear. Live sports gives brands access to large, highly engaged audiences in real time. By combining real-time sports data with first-party audience data, we help advertisers understand fan behavior and reach the right audiences in the moments that matter the most. We're seeing growing demand for the Moment Engine and our integrations across major supply side and demand side platforms allows us to capture that demand at scale.

Advertisers can tap into our data and technology through the channels that they already use, giving us broad access to advertising budgets without relying on any single platform or buying channel. The proof is in our customer growth since launching the Moment Engine. We added 174 advertisers in Q2 and another 279 in Q3, including a wide range of non-gaming brands such as Toyota, State Street, Whoop and Domino's to name a few.

Additionally, what's most encouraging is that the existing advertisers are returning to the platform and increasing their spend. In Q3, our net revenue retention was 224% compared to Q2. While the product is still in its early stages, we believe these metrics demonstrate the value that the Moment Engine can deliver. We recognize that the growth may vary over time as advertiser adoption and spending patterns evolve, but we're pleased with the early traction and remain very excited about the opportunity to build on this foundation.

Beyond helping brands make better use of existing advertising inventory, we're also creating new inventory through our data and technology. This expands the ways that we can engage sports fans and creates additional opportunities for brands to reach them.

Two developments this quarter illustrate the opportunity. First, we announced a new partnership with the NBA All-Star Chris Paul to create original athlete-led sports content powered by GeniusIQ. This is more than a content partnership. It's a powerful demonstration of how we are translating our proprietary computer vision and AI technology into differentiated fan experiences and new commercial opportunities.

GeniusIQ captures and transforms next-generation data into insights and stories that brings fans closer to the game, allowing them to see and understand basketball through the eyes of its most accomplished players. At the same time, this model creates a compelling opportunity for brands to connect with highly engaged audiences through premium athlete-led content and sponsorship. What makes this particularly exciting is the way that it brings together our technology, data and commercial capabilities to unlock new opportunities for monetization.

Second, our partnership with Univision, the largest broadcaster of Liga MX, brings a new level of insight to live broadcasters. GeniusIQ turns real-time tracking data into insights and visualizations that gives fans a deeper understanding of the game. Through our Moment Engine, these moments also become premium sponsorship opportunity, allowing brands to connect with fans directly on screen as the action unfolds.

Together, these initiatives demonstrate how we're extending the value of our data, creating more engaging experiences for fans, more valuable inventory for advertisers and additional ways to monetize our technology across the sports ecosystem.

Taking a step back, the story is clear. We've built a differentiated position across the sports ecosystem and Legend makes that position even stronger. Our data and technology underpin predictable Betting business, position us to capture growth in prediction markets and power our expanding Media business. Legend extends our reach, creates new commercial opportunities and adds further diversification of our revenue. These businesses reinforce one another, allowing us to generate more value from the same underlying infrastructure and drive greater operating leverage as we scale.

We've talked about what this business could become. We're now demonstrating what it can deliver. We're beating expectations, generating significant free cash flow and improving the earnings power of the combined company. Our strategy is delivering. Our opportunity is growing, and we're just getting started.

Before I turn to Bryan, a personal point. I'm the largest individual shareholder in Genius. In September, I sold about 800,000 shares, around 4% of my vested holding and arranged at my own cost a collar over a further 4.4 million shares. The collar sets a floor under those shares and is structured in a way that allows me to keep those shares. Maintaining those shares is of paramount importance to me because I believe in the future of this company. The cash is for my growing family and personal tax planning. You can expect me to collar further shares as that planning requires and to handle any other transactions thoughtfully in a similar vein. My intention is to unwind the collars as liquidity arise from elsewhere.

I remain substantially invested in Genius and my confidence in the business has never been higher. We've invested heavily for the future, and that investment is clearly delivering, and I'm committed to seeing it through. As a founder, I take a long-term view. As CEO, I'm accountable for meeting our commitments, generating cash and being disciplined with your capital. We will keep growing Genius and turn that growth into lasting value per share, hold us to both.

And with that, over to Bryan.

Bryan Castellani

Thanks, Mark. As we just outlined, this quarter was another strong quarter for Genius Sports. What I want to focus on in my remarks is not just the growth we're delivering, but how that growth is translating into higher margins, stronger free cash flow and an increasingly attractive financial profile. But first, let's start with revenue, where we saw continued growth across both product groups.

In Betting, revenue increased 33%, matching our strongest quarter of growth this year. This was driven by multiple factors, including a strong start to the NFL season, increased in-play betting, improving operator win margins and pricing uplift on recent contract renewals. This growth was also supported by new monetization opportunities across prediction markets, which Mark highlighted earlier.

In Media, revenue increased 126% year-over-year, which was supported by the Legend acquisition, along with underlying strength in prediction market and online sportsbook spend plus the increasing adoption of the Moment Engine.

One important point to keep in mind when assessing our Media performance is the mix of our revenue. An increasing portion of Media revenue is now generated through our self-serve platform, where revenue is recognized on a net basis. As a result, the reported revenue figure is lower than it would be under a gross presentation, but the margin profile is higher.

Importantly, we're growing group revenue significantly faster than our cost base, driving greater operating leverage and allowing more of that revenue growth to flow through to profitability. On that note, one area worth highlighting is data rights costs. The increase we reported in the first half of the year reflected the timing of new rights acquired in the fall of 2025. With those additions now behind us, cost growth is normalizing and data rights are declining significantly as a percentage of revenue. And with long-term agreements in place for our major rights, we have good visibility into these costs over the next several years and expect the rate of growth to continue declining.

We're already seeing the benefits of this operating leverage in our results. Gross profit more than doubled year-over-year with gross margin expanding to 34% from 25% in the prior year quarter.

Stock-based compensation was $15 million in the quarter, representing 6% of revenue, down from 16% a year ago. And we achieved $10 million of GAAP operating profit, which we expect to continue building on.

Finally, adjusted EBITDA margin expanded to 32%, up from 20% a year ago, reflecting the growing profitability of our combined business as we scale.

Importantly, this profitability is converting to free cash flow. This quarter demonstrates the quality of our earnings and the cash-generating power of the combined business. We converted 61% of adjusted EBITDA into free cash flow, which in line with market practice, reflects operating cash flow, less capital expenditure, capitalized software development costs and purchases of intangible assets. Excluding $16 million of cash interest expense incurred in the quarter, our unlevered free cash flow conversion was approximately 80%.

As we're discussing free cash flow and unlevered free cash flow, I want to make sure that we're clear that we are focused on free cash flow as a key financial metric going forward. For completeness and to be clear, we previously guided to over 60% unlevered free cash flow by 2028. That target remains unchanged, and we expect to progress materially towards that target in 2027. This strong cash-generating profile gives us significant financial flexibility and our capital allocation priority remains debt repayment. We are committed to delevering, and we expect net leverage to fall to approximately 2x by the end of 2026 and even further in 2027.

Moving on to guidance. The outperformance in Q3 reinforces our confidence in the business and our outlook for the full year. Therefore, we are raising our group revenue guidance to a range of $1.015 billion to $1.035 billion and adjusted EBITDA to a range of $287 million to $297 million. Additionally, we expect a year-end cash balance of approximately $260 million, representing free cash flow of more than $100 million across the second half of the year.

With strong revenue growth, expanding margins, increasing cash generation and a strengthening balance sheet, we believe our results continue to demonstrate the financial strength and scalability of our business. We enter 2027 with significant momentum, a strong foundation for continued profitable growth and increasing financial flexibility to capitalize on the opportunities ahead. We look forward to sharing more details on our 2027 outlook next quarter.

With that, I'll hand it back to Mark for closing comments.

Mark Locke

Let me close summarizing how we've delivered on our key objectives for the year. We have outperformed the guidance we set in the first 2 quarters with Legend. We reported $10 million of GAAP operating income, and we've generated $54 million of free cash flow, which underpins our confidence in our deleveraging path.

We are the infrastructure layer that our customers rely on, and we supply every regulated channel in which people bet on and watch sports. We take no view on which way the ball bounces. Our financial profile is now reflecting that. We have contracted revenue, a predictable cost base and multiple upside opportunities, resulting in significant operating leverage. This is what gives us confidence in achieving our 2028 targets of 35% adjusted EBITDA margins and at least 60% unlevered free cash flow conversion. We believe that this quarter demonstrates the progress we've made, and we look forward to updating you as we continue to execute. We will now open the line for questions.

Operator

[Operator Instructions] Your first question comes from the line of Josh Nichols with B. Riley.

分析师问答

Josh Nichols

Look, it's great to see the really strong free cash flow generation here to kick off the second half. The Betting revenue accelerated pretty meaningfully, up 33% despite some more better favorable outcomes at the start of the NFL. I'm kind of curious if you could give a little bit more granularity on what's driving that in terms of the breakdown for the renewal mix, in-play and prediction markets that's leading that growth?

Mark Locke

Yes. It's Mark. So obviously, we've talked a lot before about the way that our underlying deals have locked in revenue. So we're not particularly exposed to the wins or losses that any particular team or sport may experience at any time. So that's coming through pretty clearly in the numbers and that exposure isn't there. Prediction markets, obviously a strong growth area for us. We're seeing revenue that we make through deals with the likes of Kalshi and Polymarket, but also through the market makers as well. So we're selling to all the market makers. And just to remind everybody, we take a profit share from those market makers as well as charging a data fee in the same way that we sort of charge OSBs.

So from our point of view, the Betting business is growing particularly strong. I think you said it's 33% growth. So we're very, very happy with that. That we expect to continue to -- all the way through '27 towards '28, and we're very happy with where we are on that.

Josh Nichols

Appreciate it. And then just for my one follow-up. The Legend acquisition seems to be performing quite well. You previously talked about there's a multitude of synergies that really are not into the 2028 guidance number. If you could just provide a little bit of update on how the company has been executing on those synergies relative to plan and things like customer acquisition pricing that you're seeing today?

Mark Locke

Yes. I mean, look, the acquisition is going brilliantly. We're really happy with it. From a cultural point of view, the teams are working really closely and the business is getting fully integrated on that side. The main areas that we're focused on are obviously the commercial side. I think the evidence of that was the Kalshi and the Polymarket deals that we originally did, but there's also a number of other areas that are coming through on the commercial synergies. We're looking at doing more close work on OSBs as well as other prediction market players.

And then the other area that probably isn't so visible, but from our point of view, it's pretty exciting is the technical integration. We're bringing data in to both businesses. Prediction.com is a really good example of that. And I think the most interesting sort of -- and it's early days, but the most interesting piece of what we're doing at the moment is on the prerediction.com, where the Legend engine that we talked about a lot when we bought the deal, where we bought the business has really been sort of put out to use in that particular project, and we're seeing really strong initial results from that.

So overall, integration is going, I would say, fairly well ahead of schedule, and we think there's a lot more opportunity still to come.

Josh Nichols

Looking forward to continued progress and seeing that free cash flow KPI continue to improve over time.

Operator

Your next question comes from the line of Sam Nielsen with JPMorgan.

Samuel Nielsen

Congrats on the quarter. We've heard some other industry participants talk about increasing CPAs in the space as a result of an increasingly competitive customer acquisition environment. Obviously, we assume you sell, but wondering if you could talk about this more specifically in the scope of your performance in the quarter and maybe what your outlook is for that spend to continue going forward here?

Mark Locke

Yes, I think that's right. I mean I think it goes without saying that as the market becomes more competitive, CPAs go up. I think I've said before, we've seen CPAs double on the prediction market side since we bought Legend, which has been good. And again, we see a strong trajectory for growth over the coming time. So yes, I mean, I think what you're hearing is pretty much what we're seeing.

Samuel Nielsen

Makes sense. And then as a follow-up, appreciating that it's early and you aren't yet got into 2027, but is there any way you can kind of give us a sense of some of the puts and takes and building blocks to kind of be thinking about for next year between maybe the core business, Legend, prediction markets and maybe some offsets in Brazil or elsewhere?

Bryan Castellani

Sure. It's Bryan. We are early in planning for '27. We'll share more of that in March. As we look to the growth, it's the consistent building blocks. When you think about just the expanding TAM, both behind betting, prediction markets adds further, iGaming continuing to grow and the Legend acquisition gives us exposure to that. And so that, coupled with our continued product penetration and -- we're always adding and negotiating new deals and able to drive pricing uplift as well.

So a number of things helping drive the top line. And then as we look to margins, and you're seeing it come through this quarter and this year with operating leverage, where the cost base grows lower than the revenue base. And so you get the compounding effect and the expanding margins. And then additionally, and as you can see how we've highlighted our focus on free cash flow, that's flowing through there and continuing that ramp to our '28 guidance and building '27 will reflect that. So we're positioned well. The continued momentum sets us up and we'll share more on the next call.

Mark Locke

Yes. I think it's also probably worth adding that we've got a lot of visibility on this already because of the way the business operates and because of the deals that we've got locked in, especially on the betting side, we've got very strong -- very clear visibility of what the revenues look like, but also on the right side, what the cost base is. The rights market has been -- has sort of acted as we predicted and is coming down. So I think from our point of view, we feel increasingly confident about our '28 guide.

Operator

Your next question comes from the line of Jed Kelly with Oppenheimer.

Jed Kelly

Just going back to your earlier comments on the net revenue on Media now. I think it's pretty important to sort of look at the leverage there. I think it's now Media direct cost as a percentage of Media revenues mid-20s. How much of that is being driven by Legend versus how much of that is being driven by more people in your core fan hub experience choosing to do self-service advertising? How should we think about those margins going forward?

Mark Locke

Sure. I mean, look, we think about Legend as a component of our broader Media business. And obviously, that allows us to cover the full funnel of all the operators. CTV and video are building awareness on that at the top of that funnel, and that's powered by the sports data and of course, by the Moment Engine.

Programmatic display and native keep operating in front of those players through the consideration. So from our point of view, Legend's properties, Covers, Sportsbook Review, Oddspedia and Casino Guru reach players with the highest intent and drive the lower funnel from comparison to deposit. So we're feeling pretty good about it.

Jed Kelly

Got it. And then just as a follow-up. We are seeing market makers or smaller market makers starting to have to pay the exchanges higher fees, particularly for RFQs. Do you think this is a business model where market makers it starts to consolidate for a handful of larger scale players? And how does that impact your -- how does that impact? Is it more higher volume, but you're getting lower rev share to higher volume market makers? Just can you talk about that dynamic and the impact of higher fees?

Mark Locke

Yes. I mean the key dynamic we think about is liquidity. Obviously, the TAM in our space has expanded enormously and very rapidly with the addition of prediction markets. And I think I've said this before, Jed, the way we think about the business and the way we think about the market is about the total amount of bet, whether it's bet through OSBs or whether it's bet through prediction markets, we're fairly agnostic.

So I think the expansion that we're seeing in the TAM and the way that we make money on every bet is becoming pretty clear. The way the prediction markets evolve is going to be quite interesting. You've seen case studies in other countries with exchanges about the way that they charge. And I think we've all got a keen eye on that. But overall, from our point of view, it's just really about liquidity. The more liquidity in the market, the more money we're making is we're basically taking a slice of every bet.

Operator

Your next question comes from the line of Eric Sheridan with Goldman Sachs.

Eric Sheridan

It was great to see the level of growing advertiser adoption that you called out. Can you talk to us a little bit about both the seasonality and the structural long-term potential for growth in the advertiser pipeline and the revenue retention as well? Just so we understand some of that being either seasonal and also how you continue to possibly have confidence interval about the pipeline longer term?

Mark Locke

Yes. Thanks. Look, we're super happy with it. The strategy on Media is really about land and expand. We're looking to acquire and add in as many new customers as we can. And I think we went from 174 last quarter. We added another 279 this quarter. So we're up at sort of 450-odd customers running through that. Once we've got those customers, there's obviously a period of time where they're testing the results, making sure that we're actually delivering what we said we were going to deliver, and we're seeing net revenue retention very -- come through very well there.

Our job now is to expand the amount of money that they're spending with us. So they've got through the test campaigns. We've got a lot of customers. So it's really a case of working with those customers to drive usage over time.

Operator

Your next question comes from the line of Jordan Bender with Citizens.

Jordan Bender

One of the key themes here during third quarter has just been around some of the decelerating iGaming growth across North America. It sounds like the Legend business is performing great here. But can you just remind us how much Legend is kind of tied to market growth?

And then maybe a separate question on that is just in terms of the Legend properties, kind of what are you seeing in terms of traffic to those sites as we start the NFL season?

Mark Locke

Yes. So I mean, we -- obviously, just to remind you how we get paid on Legend, over half of Legend's money comes from a historic deal that have already been signed. So we're getting paid in some cases, from customers that were signed up 15 years ago. So the sort of annual dynamics, they're important on some basis, but not -- they don't really affect the overall macro view of Legend.

The majority of Legend customers is iGaming, as I said, but we're adding many new channels. So obviously, Covers, launch prerediction.com, some of the RIP stuff that's been coming out, I think, provides an exciting opportunity going forward. So we see fairly significant growth opportunities across the market. And as I said, from Legend's point of view, we've got a very stable, solid and reliable historical revenue stream that's delivering as we would expect.

Bryan Castellani

I just wanted to add on that, the diversity within Legend, there's diversity geographically. It's not tied to any one market or significant region. The other point is there is the Covers and the sports business. So there is iGaming and also exposure to online sports betting and prediction markets. So there's diversity within it. And then certainly, as Mark hit upon, the recurring revenue within those contracts also provides predictability as well. So less tied to any one market or any one change in the regulatory environment.

Jordan Bender

Great. And then just maybe one quick follow-up here. On the implied guidance for 4Q, how much was Brazil maybe part of the old guide that isn't in there anymore? I guess, more broadly, just how much was Brazil a percentage of revenue or how should we be thinking about that in the model?

Mark Locke

Yes. So just to remind you, the way we get paid is we have very strong contracts that don't have us exposed to any particular regulatory environment. So in the same way that we don't take risk on the outcome of sports events, we also don't take regulatory risk. So from our point of view, the closing of Brazil didn't really have any material impact on the underlying business.

Admittedly, it's probably got a little bit of -- taken some of the upside that we may have had going forward and probably muted that a little bit, but that doesn't really affect our view on any of the numbers through to '28.

Operator

Your next question comes from the line of Chad Beynon with Macquarie.

Chad Beynon

Bryan, you mentioned that leverage will be at 2x at the end of the year and then further decline here as we move forward. So given the growing free cash flow and your confidence in the earnings power, how are you guys thinking about share repurchases versus debt reduction in the near term?

Bryan Castellani

Thanks, Chad. That's right. As we grow our free cash flow and the operating leverage, the path to 2 is clear, further down in '27 and '28. We've prioritized, as we continue to say, delevering and debt repayment. We think as we get into '27, '28, we'll be well on the path of that, and that will give us optionality. But right now, the focus is that debt repayment and getting the net leverage down.

Chad Beynon

And then with respect to further BetVision or in-play adoption, why do you think you're seeing so much growth this year? The product has been out there. Is this just an evolution of the customer, kind of what they like? And how do you think this could portend for future years? Is there still upside in terms of in-play percentage from your customers?

Mark Locke

Yes. I mean, look, we've always said there'd be a shift from pre-match to in-play. I mean the benefits to customers are pretty obvious. The margins are significantly higher, and I think the customers are pretty focused on that now. So that aligns with us really well. Obviously, things like BetVision, where we've seen a significant increase in the viewing minutes on that, I think, helped change user behavior. And ultimately, the products are getting better. So I think the combination of all of those things is taking us to the place that, frankly, we sort of thought would come over time.

Operator

Your next question comes from the line of Ryan Sigdahl with Craig-Hallum.

Ryan Sigdahl

I want to start on that last question you just answered for Chad, but you mentioned kind of the benefits from an in-game standpoint and what's driving adoption and growth there, BetVision being one of those. But how do you think about that from a prediction market standpoint, evolution of the product and services and the ability to help drive even accelerated volume from the current growth they have there, namely BetVision and ways to drive in-game betting?

Mark Locke

Yes. So prediction market, I mean, already, it's about 2/3 of the volume. So it's already a very significant part. And that kind of marries with what you see in Europe in some of the more mature betting markets, or it's slightly higher to be maybe up to 70%. Obviously, prediction market product set suits in-play betting really well. And the adoption of the products of things like BetVision is something that we're clearly in conversations about and watch this space.

Ryan Sigdahl

Maybe a question for Bryan. Free cash flow was better in Q3. Kudos on that and good to see it flow through. You reaffirmed your $100-plus million target in the second half. The cash target actually implies like $105 million in the second half. You got some EBITDA upside a couple of million. It appears like that flowed to free cash flow, which is good to see. I guess, would you agree with that?

And then when you say you're going to exit the year with $260 million of cash, I guess my question is, why not proactively pay down debt quicker? Maybe it's as simple as just talking to net debt targets going forward, so you can utilize that cash faster.

Bryan Castellani

We remain consistent on the second half and generating over $100 million in free cash flow. So you saw the $54 million in Q3. Q4 will be at a comparable absolute level and sets us up well. Obviously, there can be timing across quarters. But generally, we're consistent on that second half and poised well for '27 to ramp that towards our '28 guidance.

Operator

Your next question comes from the line of Barry Jonas with Truist Securities.

Unknown Analyst

It's Jeremy on for Barry. I believe the guidance raise implies no change to Q4 assumptions. So can you just walk us through the puts and takes if there's any upside over there given the strong in-play GGR growth and new prediction market revenue contribution?

Mark Locke

It's Mark. Look, last quarter, we took it up $15 million on revenue and $15 million on EBITDA, which equates to about a 34.5% EBITDA margin for Q4. Incidentally, that's, I think, the highest Q4 margin that the business has ever had. So obviously, that's being done off the back of good momentum for prediction markets, strong integration with the Legend product. So I think we're feeling quite good about the Q4 setup. And more importantly, we're feeling very good about the setup into 2027.

Unknown Analyst

Got it. And then with NBA embracing prediction market, do you think that changes anything in terms of timing for the NFL to sign on as well? Curious your thoughts there.

Mark Locke

Look, I don't speak to the NFL. I don't know if it does or not, but all I can say is that it's not in our guidance for this year, next year or 2028.

Operator

Your next question comes from the line of Jeff Stantial with Stifel.

Jeffrey Stantial

Maybe just starting off on the Media business. So a lot going on there, Bryan, you mentioned some of the changes in self-service and how that impacts revenue growth from an accounting perspective. There's also Legend flowing in. Can you -- is there anything you could do to sort of quantify or help us think about sort of underlying organic rates of growth during Q3 for both the Legend business as well as Media as you sort of normalize out for some of the accounting changes, I guess you can look at on a gross margin basis. But just anything to sort of better frame for us sort of underlying organic growth in that business would be helpful.

Bryan Castellani

Yes. As Mark kind of walked through how we think about the business and Legend is a component of the overall and being able to work from the top of the funnel down to the bottom with operators and advertisers and marketers. We are growing that Moments Engine, just launched in March, really, really good traction and ramp coming off a small base and now up to over 400 cumulative advertisers. So good indicators there.

And you're right that the more we drive into that programmatic and curation or self-serve business that it is a net accounting where our take is lower, but our margin is higher. And so that will start to flow in more as it grows and ramps, but good traction there, and we're excited about it.

Jeffrey Stantial

Got it. And then maybe I'll stick with you, Bryan, you said something in the prepared remarks on -- I think it was that you have line of sight to moderating rate of growth in sports rights costs. Can you just expand on that point? Are your contracts structured in a way where the annual escalator steps down later on in the contract? Or is this more in regards to sort of a more rational bidding environment between you and your main player for kind of new contract come up and extensions and things of that nature?

Mark Locke

Yes, shall I pick that up? I mean, look, I mean, there's a few dynamics going on in the right space. We have a very rational market these days with us and our main competitor, which is good news. We've kind of got our lanes. And I think we're operating well in those respective lanes. I think the technology investment that we've made over the last few years is really paying significant dividends in our rights negotiations now. All of our rights deals right across the board are starting to include GeniusIQ in some way or another. And that's giving us a real position and helping the partnerships that we've got with the leagues, which again is reflecting directly in the rights fees.

So obviously, the majority of our rights fees go to the NFL and Premier League. Those are locked into the end of the decade. So we've got good visibility on that. But the rest of the market, we've got very strong position with the technology stack that we've got. And I think we feel pretty clear about the direction continuing as we've sort of outlined previously.

Bryan Castellani

The other thing I would add and just a reminder, last summer, we added a couple of rights agreements. So now that those are lapped, that also contributes to the moderating rate of growth. And as Mark said, our major contracts are locked in. We know what those are. And so line of sight and predictability gives us confidence in that.

Mark Locke

There's only -- there's one other point, which is probably worth making as well is we don't actually need any more rights deals. We've got an incredible portfolio of rights. We've got the 2 largest sports in the world for sports betting, and we've got a very strong long tail. So from our point of view, we can be very discerning about the rights deals that we do. And we've obviously got a very high bar from a profitability and accretion point of view to any deals that we would choose to do in the future.

Operator

Your next question comes from the line of Trey Bowers with Wells Fargo.

Raymond Bowers

Mark, maybe a big picture question for you. I think one question investors have in mind these days is just kind of the durability of the prediction markets. I think you did a good job kind of lining up that the whole market should continue to grow. But if we do get a Supreme Court decision sooner rather than later and the prediction market sports betting is kind of reversed, how quickly do you think the states can move? And just any sense of what that means for you guys' business, given how strong kind of the tailwinds are right now of that being replaced by more of a state-regulated environment?

Mark Locke

Yes. I mean, look, I think we, and frankly, everyone on this call probably would welcome the clarity. But I sort of stand by what we said a number of times that we believe that the market is proving its demand. There's a huge demand for prediction markets or betting liquidity to be placed somewhere in the market. And at the moment, that's through betting on the prediction market.

But ultimately, the thing that I keep coming back to is I think the thing that's going to change is the regulatory wrapper around it. You've got -- do I think that prediction markets will look exactly like they do in 3 years' time as they do today? I think it will evolve. But I think what is going to be durable is that betting volume. We've seen it in other countries.

And I think what -- the opportunity for us through the TAM expansion is really quite significant because not only is this a catalyst for sort of growth in our business, but I think it's a catalyst for states to change. Once you -- if you've seen there's a number of states that are already taxing prediction markets, they're looking at ways of working with them. You've got the tribes who have already indicated that they would like to regulate betting space to move forward. You're seeing those sort of dynamics, which I think are very beneficial.

And ultimately, if you look at it from a taxation point of view, you can now quantify the taxation level that these states will be turning down because they know how much prediction markets are generating. So I think the argument is a very different argument to, one, that's previously been where everyone has been slightly blind on this. You know how much money is going through these markets and you know what the taxation rates could be. So I think it's very -- it becomes a very different conversation.

I think the other thing that's going on is it's sort of bipartisan issue. It used to be, I think, something that I think the Republicans were pushing pretty hard, but Kalshi and Polymarket have done a very good job of making that bipartisan issue with some of the hires and some of the work they've done on the political side. So I think all of these benefit pretty well the sort of durability of the markets going forward.

Raymond Bowers

Awesome. And then a quick follow-up question, Bryan. Great to see the free cash flow really turn a corner this quarter and expected next. You provided the 60% unlevered free cash target by 2028. Could you just give -- I think you've maybe given this before, but could you remind us what the interest expense or leverage assumption against that is just so we could kind of bridge from unlevered to free cash conversion by '28?

Bryan Castellani

Yes. On an unlevered basis, as we've said, it's over 60%. Our interest is between $50 million to $60 million a year. And so that over 60%, I would tell you, becomes over 50% on a free cash flow basis.

Operator

Your next question comes from the line of Greg Gibas with Northland Securities.

Gregory Gibas

Congrats on the results. I know you likely don't want to speak for them, but to the degree that you can share what you're hearing from the NFL as it relates to their updated stance on prediction markets and if anything has changed there from your perspective?

Mark Locke

Look, I can't speak -- I don't speak for the NFL. I can't speak for the NFL you're right, I don't want to speak to the NFL. Look, I think the biggest clues in the market that you should look at are what's going on in the other sports and some of the conversations that have been sort of made public that the NFL have put out there. I think, as I said in sort of last comments, we're going to welcome some clarity with the Supreme Court. But from Genius' point of view and bringing it back to what we do, we haven't assumed it in any of the numbers that we have put out there and what we think is going to happen at any point. And our guide for '28 stands without the NFL.

Gregory Gibas

Got it. Fair enough. And I wanted to follow up on a previous question, but as it may be related to the upside in relation to your prior expectations or your expectations going in, was it mostly prediction market and/or market maker derived versus OSB derived based on GGR growth? Or would you say just generally broad-based?

Mark Locke

Yes. I mean, look, it's broad-based and it is Legend. We're seeing strong growth across the business. Legend is integrating really well. We're super happy with the product sets that we're putting out. The market dynamics are good for us as well. There's some tailwinds. So we -- overall, the business is performing, frankly, as we expected it to.

Operator

We have reached the end of the Q&A session. This concludes today's call. Thank you for attending. You may now disconnect.