Fox buys Roku for $22bn as UK publishers bill AI scrapers
Fox buys Roku for $22 billion, reshaping CTV advertising consolidation, as UK publishers bill AI scrapers £500 per article through county court contracts now.
Fox buys Roku for $22 billion, reshaping CTV advertising consolidation, as UK publishers bill AI scrapers £500 per article through county court contracts now.
The most consequential deal in connected television advertising in years landed on June 15, 2026. Fox Corporation agreed to acquire Roku for approximately $22 billion in enterprise value, in a cash and stock transaction that values each Roku share at $160. Fox will pay $96 per share in cash and 0.9693 shares of Fox Class A common stock for each Roku Class A and Class B share outstanding at closing. That represents a 34 percent premium to Roku’s unaffected share price as of June 11, 2026, and a 21 percent premium to Roku’s unaffected 52-week high. The transaction is expected to close in the first half of 2027, pending shareholder votes and regulatory clearance under the Hart-Scott-Rodino Antitrust Improvements Act.
The strategic logic is vertical integration at a moment when the converged television advertising market is consolidating rapidly. Fox already owns Tubi, the ad-supported video-on-demand service it acquired in March 2020 for $440 million. At the time, Fox sold its stake in Roku to fund that purchase. What followed was six years of parallel growth: Tubi became one of the most-watched free streaming services in the United States, while Roku built a platform reaching more than 100 million households globally and a supply-side ad stack centred on its OneView demand-side platform. The June 15 deal brings both sides back under one owner.
The combined company will become the third-largest player in US television by share of viewing, sitting behind only YouTube and Netflix in streaming reach. Tubi and The Roku Channel together accounted for 5.2 percent of all US streaming viewership as of March 2026, per Nielsen Gauge data. Fox already held formidable live sports rights across NFL, MLB, NASCAR, and college football. Adding Roku’s homescreen technology, first-party audience data from 100 million global streaming households, and The Roku Channel’s FAST inventory creates a stack that covers the entire funnel: awareness through live sports and news, consideration through on-demand content on Tubi, and performance through Roku’s OneView performance marketing tools, including a data partnership that allows targeting using Amazon e-commerce behaviour.
Lachlan Murdoch, executive chair and chief executive of Fox Corporation, described the rationale on a call with analysts on June 15. The combination creates a platform at the intersection of live sports and streaming, with Roku’s first-party data and performance marketing capabilities layered on top. Roku chief executive Anthony Wood is expected to join Fox’s board following the transaction close. Both companies committed in their SEC filings to keeping Roku operating as an open, partner-friendly platform that continues to distribute competing streaming services.
That neutrality pledge will face scrutiny. When a competitor owns the operating system and homescreen through which third-party streaming services reach audiences, the incentive structure changes. Promotional placement, default positioning, and algorithmic discovery are all levers that a platform owner can adjust. Whether regulatory conditions attached to the deal will constrain those levers remains to be seen.
For the programmatic advertising market, the deal compresses the number of independent sell-side operators in converged TV. The pending Paramount and Warner Bros. Discovery merger was already reducing that field. Fox adding Roku means one seller now controls homescreen advertising inventory, FAST channel distribution, AVOD inventory through Tubi, and the live sports rights that command the highest CPMs in television. PPC Land reported the deal’s advertising implications in detail on June 15, noting that the combination gives Fox a unified first-party data asset of unusual scale and a set of performance marketing tools that most broadcast networks cannot match.
The deal did not arrive without warning. PPC Land reported on June 13 that Roku was exploring strategic options including a full sale, with its 100 million-household advertising platform drawing interest from multiple media and technology buyers. Reuters had reported that Roku executives were in active negotiations. The $22 billion figure that materialised sits well above the $19 billion figure that had circulated in early reports, reflecting competitive tension in the acquisition process.
Related CTV developments on the same day reinforced the broader picture. VAB published June 2026 streaming data showing that 89 percent of US streamers now use ad-supported services, that FAST channels had reached 1,700 titles across available platforms, and that interactive CTV ads produce 138 percent higher brand recall than standard pre-roll. Those figures describe the environment Fox is buying into: ad-supported streaming has become structurally dominant, not a niche preference. Roku’s OneView platform is built for exactly these formats. Fox will inherit that infrastructure at the moment of peak consumer shift toward ad-supported content.
In Europe, a parallel but smaller-scale dynamic played out on the same day. Zattoo named Ströer as its exclusive ad sales partner for German CTV and video inventory, covering audiences across smart TVs, streaming players, and mobile devices. European CTV is rationalising through commercial partnerships rather than acquisitions, with publishers outsourcing monetisation to specialist sales houses. How long that model remains viable if US-style consolidation accelerates into European streaming markets is an open question.
Earlier in the week, Gray Media, the largest US local television station owner by number of markets, partnered with Madhive’s AI-powered DSP and its Maverick platform to target audiences across 117 markets, timed for the 2026 midterm election ad cycle. Political CTV spending is projected to reach $2.7 billion for the cycle. Local broadcasters watching a Fox-Roku combination consolidate national CTV inventory will face increased pressure to automate their own ad sales infrastructure or seek their own partnership arrangements.
UK publishers bill AI scrapers through the county court
A group of UK publishers launched Search-Only Contracts on June 15, 2026, a legal mechanism designed to charge AI companies including OpenAI and Google £500 per article scraped, with enforcement pursued through county courts rather than through intellectual property litigation. The approach is significant precisely because it bypasses the cost and complexity of IP law.
The mechanism rests on a contractual argument. When an AI crawler accesses a page carrying terms that explicitly grant access only for the purposes of appearing in search results, any use beyond that, including training large language models or powering AI-generated answers, constitutes a breach of those terms. The £500 per-article figure is calibrated to the county court small claims track, which avoids the legal costs that make traditional IP enforcement prohibitive for smaller publishers. Enforcement through the small claims track does not require specialist intellectual property counsel, which is the practical barrier that has kept most individual publishers from pursuing AI operators over scraping.
This connects directly to the legislative track running in parallel. New York’s Assembly passed bill A11292 on June 5, 2026, requiring AI crawlers to disclose their identity and purpose to news publishers or face penalties of $15,000 per day of non-compliance, as reported by PPC Land on June 14. That bill targets the disclosure problem: publishers currently cannot reliably distinguish between legitimate search crawlers, which generate referral traffic, and AI training crawlers, which consume content without generating any return. A11292 would require AI operators to provide that distinction in machine-readable form.
The UK contractual route and the New York disclosure legislation are two different interventions in the same underlying problem. AI companies are consuming publisher content at scale without compensation, while simultaneously redirecting the user traffic that the content would otherwise generate. The traffic destruction is measurable and documented. PPC Land reported extensively on this on June 14. All About Berlin, an independent guide to Berlin immigration that had operated for eight years, lost 70 percent of its organic search traffic to Google AI Overviews, with its founder reporting the decline directly. Overfishing.org, a 21-year-old independent site on marine conservation, shut down entirely after AI Overviews absorbed its organic search traffic.
These cases illustrate the asymmetry at the heart of the dispute. Publishers produce specialised, fact-dense content that AI systems find valuable for training and for answering user queries in real time. The AI system then provides that answer directly, reducing the probability that a user clicks through to the original source. Traffic falls. Revenue falls. The capacity to produce future content falls. The AI system’s training corpus was built on content that no longer has a viable business model to sustain its own creation. PPC Land’s AI Overviews and ad tech trust seals newsletter from June 15 described this cycle explicitly, noting that the traffic destruction had produced real site closures, not just traffic declines.
The UK Search-Only Contracts may face enforcement friction. County courts are equipped for straightforward debt claims. Establishing that a particular article was scraped by a specific crawler at a specific date requires technical log documentation that most publishers do not routinely collect with sufficient granularity for litigation. The £500 figure is also modest relative to the scale of content AI systems consume, though it is designed to function at volume: a publisher with several thousand scraped articles could theoretically pursue five-figure claims without specialist legal representation.
Perplexity’s situation added another dimension to the same week’s coverage. PPC Land reported on June 15 that Perplexity co-authored a study claiming AI agents cut task time by 87 percent, while copyright and security cases continued to mount against the company. The study used Perplexity’s own data and products to validate Perplexity’s own performance claims. That self-referential methodology drew attention to a credibility problem that sits at the centre of the AI content dispute: AI companies are simultaneously the subject of the legal complaints and the producers of the evidence used to justify their operations.
Google Ads prepares for a three-part overhaul in August
Google announced on June 15, 2026, three significant changes to its Google Ads platform, all scheduled to take effect on or around August 17, 2026. PPC Land reported all three updates with detail on the mechanics of each change.
The first is the expansion of Smart Bidding Exploration to Performance Max campaigns. Smart Bidding Exploration is a bidding mode that actively tests outside an advertiser’s current targets, looking for conversion opportunities the existing strategy would not reach. Applying this to Performance Max campaigns, which already operate with substantial automation and limited placement transparency, compounds the opacity. Advertisers who rely on target CPA or ROAS settings to constrain Performance Max behaviour will find that exploration mode actively tests outside those parameters. The feature had previously been limited to search and shopping campaigns.
The second change is a promotion mode beta. This gives advertisers a dedicated interface for setting promotional parameters on campaigns, separating promotional logic from the base bidding strategy. Rather than encoding promotional periods through bid adjustments or temporary target changes, advertisers can set promotion-specific parameters that activate and deactivate on a schedule. The beta launches in the United States first.
The third change, and the one that drew most attention from practitioners on June 15, is a modification to bidding target optimisation. Google framed it as delivering more predictable performance for campaigns with fixed spending limits. The language is precise but ambiguous. Fixed spending limits constrain the budget envelope, but they do not constrain how Google allocates bids within that envelope across placements, devices, and audiences. Advertisers who have interpreted target CPA or ROAS as hard constraints may find those targets treated as optimisation parameters rather than firm limits after August 17.
Search Engine Roundtable covered the same three updates on June 15, with Barry Schwartz flagging the bidding target change as the most significant. Schwartz noted that Google described the change as improving predictable performance, but the practical effect on accounts with tight target constraints will depend on implementation details that Google has not yet published.
Google also confirmed on June 15 that it was expanding its limited ad serving policy to cover Google Search, as a phased enforcement programme running from June 2026 through 2028. PPC Land had reported this on June 13. The policy targets what Google calls unqualified advertisers, those whose accounts show patterns associated with negative ad experiences for users. Affected accounts will see reduced ad impression volume rather than account suspension; the policy is graduated rather than binary.
Google’s background Search agents expanded on June 13 to all AI Mode languages and markets, restricted to Google AI Ultra subscribers. Search agents were announced at Google I/O 2026 and allow the search engine to perform tasks and retrieve information in the background, even when users are not actively querying. The combination of search agents, AI Mode, and the August bidding changes represents the most concentrated set of Search platform updates Google has shipped in a compressed period since the introduction of Smart Bidding itself.
The timing overlaps with Google’s delayed DSA-to-AI Max migration. PPC Land reported on June 13 that Google pushed the automigration deadline from September 2026 to February 2027, restoring DSA campaign creation in the interim. Advertisers who were managing against the September deadline now have an extended runway, but the February 2027 date is firm. The August bidding changes arrive while that migration is still being planned across the industry.
Microsoft Advertising brings LinkedIn seniority targeting to 29 markets
Microsoft Advertising announced on June 15, 2026, that advertisers can now filter search and audience campaigns by job seniority using LinkedIn profile data, covering 10 seniority levels across 29 markets including the United States and EMEA territories.
The 10 levels span unpaid and training positions through senior individual contributor and director-level roles up to C-suite and partner-level seniority. The 29-market rollout makes this the broadest geographical expansion of LinkedIn-derived seniority targeting that Microsoft Advertising has offered. The existing LinkedIn profile targeting suite already includes job function, industry, and company size. Seniority adds a fourth firmographic dimension with significant implications for B2B search campaigns.
The practical use case is intent-plus-firmographic filtering. A technology vendor bidding on the query “enterprise software procurement” can now weight bids toward users whose LinkedIn profiles indicate director-level or above seniority, while reducing or eliminating spend on the same query from users in junior or training roles who are unlikely to be involved in purchasing decisions. Previously, advertisers could filter by job function or industry but not by the seniority of the person conducting the search. Seniority is often the variable that determines whether a search leads to a qualified sales conversation.
Context from HubSpot’s projections on June 15 is relevant. PPC Land reported that IDC data projected $42 billion in HubSpot partner revenue by 2030, with AI-first revenue growing at a 28.4 percent compound annual growth rate. A separate HubSpot survey of more than 3,000 buyers, reported by PPC Land on June 13, found that AI search had become the single strongest predictor of CRM purchase intent, ahead of product demos, peer reviews, and direct sales calls. Together those two findings describe a B2B buying journey that is bifurcating: AI-assisted discovery dominates the top of the funnel, while identity-linked advertising platforms like Microsoft’s LinkedIn integration become the mechanism for reaching specific decision-makers once intent has been established.
Uber Advertising extends beyond its own platform for the first time
Uber Advertising announced on June 15, 2026, three new products: Offers on Uber, placing sponsored incentives in the rideshare booking flow; expanded premium formats on Uber Eats including brand takeover placements on the app’s home screen; and Offsite Ads, which extend Uber’s audience targeting beyond the Uber ecosystem to third-party inventory for the first time.
Offsite Ads is the most structurally significant of the three. Until June 15, Uber Advertising had operated entirely within its own surfaces, the rideshare app and the Eats ordering app. Offsite Ads creates an audience export mechanism allowing brands to reach Uber users on external placements using Uber’s first-party data on mobility behaviour, food ordering patterns, and location history. That is the retail media model applied to a mobility platform: using transactional first-party data to power advertising well beyond the surface where the data was originally generated.
The pattern maps directly onto the broader retail media measurement problem. PPC Land reported on June 15 on a meta-study by Incremental covering 150,000 campaigns and $350 million in ad spend, which found that siloed attribution in retail media misses between 36 percent and 53 percent of true campaign ROI. The implication for Uber’s Offsite Ads product is immediate: conversions that result from Uber audience targeting on third-party inventory will appear in those third parties’ attribution systems, not in Uber’s, unless advertisers build specific cross-channel measurement frameworks to capture the full effect.
The launch follows Uber Advertising’s June 8, 2026, introduction of Deal Drops and Reorder Rewards on Uber Eats, reported by PPC Land on June 14. Deal Drops are time-bound event offers that restaurant advertisers can activate around specific occasions. Reorder Rewards are post-purchase mechanics designed to drive repeat ordering from customers who have already converted. Together, the June 8 and June 15 launches outline a full-funnel advertising stack: awareness through Offsite Ads and brand takeovers, consideration through in-app placement, and loyalty mechanics through Reorder Rewards.
Uber’s expansion into offsite audience targeting arrives as the retail media market faces measurement pressure from multiple directions. Commerce media leaders are debating how to value AI recommendation visibility: Koddi research cited by PPC Land on June 14 found that 84 percent of commerce media leaders would invest in AI recommendation visibility, with 70 percent already using agents for campaign execution. Uber’s first-party mobility and food ordering data has clear relevance for those recommendation systems, and the Offsite Ads product may eventually serve as a bridge between traditional programmatic placements and AI agent-mediated recommendation surfaces.
Also noted
June 15, 2026: DeepIntent launched Helix AI, described as the first agentic platform for healthcare marketers, claiming to cut media planning cycles in half through HIPAA-compliant natural language tools designed for pharma and health brands.
June 14, 2026: HUMAN Security data for May 2026 showed agentic AI traffic fell 4.3 percent month over month while publisher blocking rates rose to nearly 9 percent, as Comet, Atlas, and Claude compete for agent traffic share across publisher properties.
June 15, 2026: Brainlabs Chief Growth Officer Ali Reed argued that independent agencies are capturing the high-margin, AI-driven work that legacy holding companies are ceding to automation, describing the structural shift as a window that will close once the holdcos adapt.
June 14, 2026: TAG’s fraud-prevention certifications for Google and The Trade Desk lapsed, while LinkedIn invalid traffic reached 17.62 percent in Q1 2026 per Lunio’s 64 million-click analysis, and political CTV spend projections rose to $2.7 billion for the 2026 midterm cycle.
June 14, 2026: VAB’s 2026 addressable TV guide showed 92 percent of pay TV households are addressable-enabled, with nearly half of current addressable advertisers raising their budgets in 2026, a number that will likely accelerate now that Fox controls Roku’s addressable infrastructure.