Private equity firm Vista Equity Partners and its partner Quinti Capital submitted a takeover offer for Criteo last week worth more than 50 percent above the commerce media platform's recent share price, two sources familiar with the matter told Reuters. The disclosure, which first surfaced through Bloomberg News on Monday, July 6, sent Nasdaq-listed shares of the Paris-based firm up 21.4 percent to close at 23.17 dollars, giving the company a market value of roughly 1.16 billion dollars. Criteo has not yet decided how it will respond. The bid did not arrive in isolation. It landed in the same week Google began a six-week countdown toward a bidding change that will push many advertisers' campaigns toward higher stated costs, the same week the Federal Trade Commission opened a comment period on a proposal that could preempt state AI bias laws, and the same day a media measurement firm circulated benchmark data showing fraud rates falling even as geo-targeting accuracy worsens. Taken together, the last day of coverage across PPC Land, Digiday, AdExchanger, Search Engine Roundtable, Adweek and MediaPost tells a story about an industry absorbing simultaneous shocks from private capital, platform mechanics, federal regulators and its own measurement infrastructure. None of the developments below happened by coincidence in isolation from one another; each connects, in some fashion, to questions the others raise about who controls advertising infrastructure and on what terms.

Criteo, a company mid-transition, becomes a takeover target

Criteo did not arrive at this moment as a stable business quietly attracting outside interest. It arrived as a company absorbing a self-inflicted wound while simultaneously restructuring its legal domicile to make an acquisition easier to execute, a combination that helps explain both the timing of the bid and the market's swift, credible reaction to it.

The wound traces back to May 2, 2025, when Criteo disclosed that its largest retail media client would discontinue managed services and curtail brand demand sales beginning that November, PPC Land reported. The company estimated the change would cost 25 million dollars in 2025 and as much as 75 million dollars across the first ten months of 2026. That pressure showed up repeatedly in subsequent quarters. Fourth-quarter 2025 revenue fell 2 percent year-over-year to 541 million dollars, with a 25 million dollar hit tied directly to the scope reduction. First-quarter 2026 results, announced May 6, showed GAAP revenue down 6 percent to 424.6 million dollars and net income down 79 percent, even as total media spend crossed 1 billion dollars in a single quarter for the first time, a divergence that illustrates how Criteo's platform-based model is shifting the relationship between the dollar volume it facilitates and the revenue it recognizes directly.

Alongside that earnings pressure, Criteo has been engineering a structural change that would make a transaction considerably easier to execute. On October 29, 2025, the company announced plans to redomicile from France to Luxembourg, citing a specific legal barrier: French law provides no framework for a direct merger into a US corporation, while Luxembourg's cross-border merger regime carries no such limitation. Shareholders approved the conversion by an overwhelming margin at an extraordinary general meeting on February 27, 2026, casting 50,511,371 votes in favor against just 114,993 opposed. Frederik Van der Kooi, Chairman of Criteo's Board, said at the time that the vote set the company on course to become a Luxembourg entity in the third quarter of 2026, increasing strategic flexibility. That redomiciliation window overlaps precisely with the week Vista Equity Partners and Quinti Capital submitted their offer, whether or not the conversion had formally completed by the time the bid landed.

Why would a private equity buyer see an opening in a company navigating both revenue pressure and a legal restructuring simultaneously? Part of the answer lies in what Criteo has built alongside the difficulty. On February 5, 2026, the company introduced its Agentic Commerce Recommendation Service, reporting up to 60 percent improvement in recommendation relevancy compared with text-based approaches, based on internal testing conducted that January. Less than a month later, on March 2, Criteo became the first advertising technology partner in OpenAI's ChatGPT advertising pilot, connecting roughly 17,000 advertiser clients to ChatGPT's Free and Go subscription tiers in the United States. By early May, more than 1,000 brands were running active campaigns through that integration, with AI-referred conversion rates approaching twice the rate of traditional search referrals in some retail categories. MediaPost's Laurie Sullivan, writing on July 6 about Criteo's ad business and partner network, framed the company's multi-year buildout in similar terms: years of assembling infrastructure now sit inside a company whose public market valuation reflects near-term revenue weakness rather than the platform's longer-run positioning.

Vista Equity Partners brings its own recent history in adtech to the table. The Texas-based firm sold its stake in Integral Ad Science to Novacap in a 1.9 billion dollar all-cash deal announced in September 2025, and Vista executives Rod Aliabadi and Eric Roza have held board and operating roles at TripleLift, another adtech company in the firm's portfolio. Vista's own company materials describe a public roster spanning more than 90 companies serving over 450 million users worldwide as of December 31, 2025, with holdings concentrated in enterprise software categories including identity management, cybersecurity and fintech alongside a smaller cluster of marketing technology names. That breadth is a defining feature of the firm's approach: it buys software businesses carrying recurring revenue, works to improve their operations over a period of years, then sells or takes them public again, a pattern that fits the profile of a company like Criteo whose core business has slowed even as its technology platform retains clear strategic value.

That Novacap connection produced an unusual coincidence on the very day the Criteo bid became public knowledge. Adweek reported that Integral Ad Science named Lidiane Jones as Chief Executive Officer effective immediately on July 7, succeeding Lisa Utzschneider, who led the company for more than seven years and will remain as Special Advisor to the Board through the end of 2026. Jones previously served as CEO of Bumble and CEO of Slack following its acquisition by Salesforce, and spent more than a decade earlier in her career at Microsoft. Utzschneider will also serve as a Special Advisor to Novacap and its portfolio companies, tying the leadership transition directly back to the same private equity ecosystem now circling Criteo. Whether that is coincidence or a signal of broader repositioning inside Novacap's adtech holdings is not something either company has addressed publicly, but the overlap is difficult to ignore given how recently Vista exited its own IAS stake.

MediaPost independently confirmed the takeover interest under the headline "Criteo Attracts Takeover Bids From Equity, Investment Firms," corroborating the Reuters and Bloomberg reporting and suggesting the story had, within hours, become one the wider trade press treated as settled fact rather than speculation. Several details remain genuinely unsettled: whether the offer is all-cash, whether it involves the kind of debt financing typical of leveraged buyouts, and whether Criteo's board has formally engaged financial or legal advisors to evaluate the approach. Reuters' sources described the offer as submitted "last week" without specifying an exact date, and none of the three parties involved had issued a public statement confirming terms as of Tuesday. Criteo's own operations, meanwhile, continue running on two segments regardless of how the takeover question resolves. Retail Media covers sponsored product placements and display formats appearing directly on retailer websites and apps, a business the company runs for roughly 225 retailers globally, while Performance Media, the older of the two segments, is built on personalized retargeting across the open web, the product line that originally established the company after its 2005 founding. Combined, the platform serves more than 4,100 brands, a scale that means any change in ownership carries stakes well beyond the headline valuation figure, since Criteo functions as commerce infrastructure for a large number of retailers and brands rather than as a single-purpose media buying tool.

The deal, if it proceeds, would be one of the larger adtech ownership changes of 2026 in a year already marked by unusually active consolidation. Publicis Groupe announced on May 17 an agreement to acquire LiveRamp for a total enterprise value of 2.2 billion dollars, a 29.8 percent premium to LiveRamp's closing price, a deal that has generated debate about what happens when a shared identity-resolution utility passes into the ownership of a single agency holding company. That same LiveRamp now finds itself running a five-month brand campaign on Netflix, Adweek reported, promoting its "steadfast trustworthiness" around media neutrality and data security during commercial breaks aimed at business executives who are streaming rather than reading trade press. AdExchanger's framing was pointed: LiveRamp must simultaneously reassure clients being courted by rivals with worst-case scenarios and satisfy global privacy and antitrust regulators who have not yet approved the Publicis deal. A company defending its neutrality through a TV ad campaign, while a private equity consortium bids on a rival platform in the same product category, captures something true about where adtech ownership currently sits: consolidating fast enough that the companies involved feel compelled to make the case for their own independence in primetime.

Google reshapes bidding as a six-week clock runs down

Away from the deal-making, Google Ads introduced a mechanical change to its bidding systems that will alter, for a large number of advertisers, the actual cost figures showing up in campaign reports come mid-August. The Bid Target Adjustment Tool became available inside Google Ads accounts on July 6, Google's own Help Center documentation confirms, arriving roughly six weeks ahead of the date it exists to help advertisers prepare for. PPC Land's reporting on the tool's launch details how the underlying change takes effect August 17, when campaigns carrying a "Limited by budget" status and running Target CPA or Target ROAS bidding will stop being permitted to quietly outperform their stated targets.

The mechanics, as Google has laid them out, are specific rather than vague. A campaign with a Target CPA of 10 dollars that has actually been converting closer to 5 dollars for months will not continue pocketing that gap after August 17; delivery will instead move toward the 10-dollar figure the advertiser originally entered. The same logic runs in reverse for Target ROAS: a campaign set to a 200 percent return that has been running closer to 400 percent will drift back toward the lower, stated figure once the change takes hold. Google frames the update as a fix for a longstanding inconsistency, since budget-limited campaigns using target-based bidding have historically been able to overperform their targets while producing unpredictable swings whenever a budget was adjusted. After August 17, the company states, delivery will track more consistently toward whatever figure an advertiser actually typed in, whether that figure reflects current reality or not.

Eligibility spans a wide set of formats. Search, Shopping, Performance Max, Demand Gen and Travel campaigns are all included in the August 17 change. Hotel and Display campaigns already operate under the new behavior and will see no shift on that date, while Google's documentation excludes App Campaigns, Video reach campaigns and Video view campaigns from this update entirely. Platform coverage extends beyond the standard interface to Search Ads 360, Display and Video 360, Google Ads Editor and the Google Ads API, meaning agencies managing accounts through enterprise tooling face the same deadline as advertisers working directly in the browser.

Search Engine Roundtable independently corroborated the substance of the change on the same day, publishing the full text of Google's guidance and, notably, an actual screenshot of the notification email Google sent to advertisers. Barry Schwartz's coverage cited advertiser Arpan Banerjee, who posted the email he received directly to LinkedIn, giving practitioners a first-hand artifact of what the notification looks like inside a real account rather than relying solely on Google's published documentation. The Search Engine Roundtable piece also captured Google's framing verbatim: "Starting August 17, 2026, Google is updating its bidding systems to deliver more predictable campaign performance," applying to "Search, Shopping, Performance Max, and Demand Gen campaigns that use a target-based bid strategy."

Inside the Bid Target Adjustment Tool, Google lays out three distinct paths for advertisers reviewing an affected campaign. An advertiser can keep the existing target unchanged, accepting that the campaign will adjust toward that stated figure after August 17. An advertiser can lower the target to match recent actual performance, effectively locking in current efficiency before the platform pulls delivery back toward the old number. Or an advertiser can set an entirely custom target reflecting updated business goals, distinct from both the original figure and recent performance. Google's separate guidance on Smart Bidding notes that the system reacts to target changes within minutes but can take one to two full conversion cycles, often around seven days, to actually settle at the new target, and warns against making multiple changes within a single cycle since doing so gives the bidding system, in the company's own words, "multiple versions of the desired outcome," which impedes performance rather than improving it.

Reaction inside the practitioner community, captured in a LinkedIn thread PPC Land documented, split between resigned acceptance and open skepticism about Google's motives. One performance marketer characterized the update as another example of Google squeezing additional profit from advertisers who had grown comfortable with favorable, unadjusted targets. A veteran digital marketing leader offered a more measured reading, suggesting the change should produce more predictable behavior over the long term while acknowledging the transition itself would likely prove tricky in practice. A third practitioner, a paid search advisor with prior platform-side experience, called the update a "money grab" in blunter terms, while another agreed the framing was "odd" without elaborating further. Both readings, PPC Land's coverage noted, can be true simultaneously: the system genuinely will behave more predictably once budget-limited campaigns consistently deliver toward stated targets, and advertisers who have not reviewed their targets recently, treating a favorable CPA as a stable baseline rather than a signal their stated target no longer matched reality, will very likely see reported costs rise unless they intervene first.

Complicating the picture further is an unrelated cosmetic change rolling out in the same window. Starting in June 2026, Google began relabeling how bidding strategies display inside the interface, with "Maximize conversions with a Target CPA" becoming simply "Target CPA" and "Maximize conversion value with a Target ROAS" becoming "Target ROAS." Google is explicit that the two labels function identically and that the change is purely visual, but the shared timing means an advertiser checking account settings during this stretch of weeks could reasonably wonder whether the renamed label connects to the performance shift due on August 17. It does not. One is cosmetic. The other rewrites how the algorithm treats a budget-limited campaign's stated cost target, and Google's own documentation treats the two as separate initiatives even while cross-referencing each from the other.

A federal challenge to state AI rules names Colorado directly

While advertisers absorbed a mechanical platform change, a separate development out of Washington raised a more structural question about how AI systems built for advertising, search and customer engagement might need to operate under diverging federal and state expectations. The Federal Trade Commission released a proposed policy statement on July 1 arguing that AI companies which secretly steer their systems' outputs toward undisclosed ideological objectives may be violating Section 5 of the FTC Act, PPC Land reported. The nine-page document, approved by a 2-0 Commission vote and published on public inspection at the Federal Register on July 6, opened a comment period running through July 31. It names Colorado's revised Artificial Intelligence Act, most recently amended through Senate Bill 26-189, as a law that could pressure AI developers into exactly the kind of concealed manipulation the statement targets.

The core legal argument extends a nearly ninety-year-old framework to a new category of product. Section 5 prohibits unfair or deceptive acts in commerce, and the Commission has applied a consistent three-part deception test since 1984: a representation, omission or practice likely to mislead a reasonable consumer, where that misrepresentation is material to the consumer's decisions. The statement itself argues that AI companies have marketed their products for years as tools designed to produce the most accurate output possible within their technical limits, and that consumers have absorbed that message, trusting AI systems on medical questions, financial guidance and ordinary homework help. The document treats that marketing itself as the source of a legal obligation. If a company advertises its system as pursuing the best answer and instead quietly optimizes for something else, without disclosing that trade-off, the mismatch between promise and product can become deceptive under the same framework the FTC has used against fake reviews and undisclosed paid endorsements.

The statement takes care to separate two very different failure modes. Ordinary hallucinations caused by a model's technical or resource limitations sit outside its target entirely; nothing in the document treats an occasional factual error as illegal by itself. What the Commission targets instead is a deliberate design choice: training or configuring a model to prioritize an undisclosed objective ahead of accuracy while continuing to market the system as accuracy-first. One passage in the statement describes a hypothetical scenario in which a company trains its model to correct what its own developers consider "historical injustices" in the facts, without disclosing that override to users, framing it as a textbook example of the conduct the policy targets.

Colorado's law enters the picture because the statement argues a state requirement banning "algorithmic discrimination" could force AI developers to distort otherwise accurate outputs to avoid liability for statistically uneven results across protected groups. If a company makes that adjustment quietly while continuing to market its system as accuracy-first, the statement contends the company has deceived consumers regardless of whether the underlying motive was state law compliance. This is not the first time federal action has singled out Colorado specifically. Executive Order 14365, signed by President Trump on December 11, 2025, directed the FTC to prepare guidance on how state AI rules interact with Section 5, and the US Department of Justice filed a complaint in intervention on April 24, 2026, joining a federal lawsuit brought by xAI against Colorado's Attorney General that argued the state's original statute compelled discrimination in violation of the Equal Protection Clause. A federal court blocked enforcement of that original law on April 27, 2026, prompting Colorado's legislature to repeal and replace it with SB 26-189, which cleared the state House 57-6 on May 9 and now carries a January 1, 2027 effective date. A footnote in the new FTC statement makes clear the revised law still concerns federal regulators, noting Colorado "materially revised" its statute but that the new version "poses many of the same concerns" as the original.

Several elements remain genuinely unresolved. The statement does not define precisely where routine safety fine-tuning, which nearly every AI developer already documents in some form, ends and prohibited "ideological steering" begins. The preemption theory it advances is untested in court, and federal judges have not yet ruled on whether Section 5's deception framework actually displaces a state consumer protection statute like Colorado's. The document itself acknowledges the FTC Act does not expressly preempt state law, resting its argument instead on implied conflict preemption, a doctrine requiring courts to find it genuinely impossible to comply with both state and federal requirements at once.

The relevance to marketers extends beyond any single company's compliance calendar. Practical friction between AI-driven advertising tools and their real-world behavior surfaced the same week in far more mundane form. MediaPost's Laurie Sullivan documented several concrete glitches inside OpenAI's ChatGPT advertising rollout, reported by Anthony Higman, founder and CEO of paid media agency Adsquire, over the two days prior. Higman described watching ChatGPT pull down an entire ad group and send every ad back into review after an advertiser accepted one of the platform's own suggested ad variations, and separately described a client whose Google AI Mode listing served correctly for the search term "Bankruptcy Lawyers Near Me," only for the AI-generated answer to attach its citation link to a competitor's website rather than the business that had actually ranked. Neither incident touches the FTC's ideological-steering theory directly, but both illustrate the same underlying vulnerability the Commission's statement is reaching toward: AI systems making representations, whether about accuracy, attribution or intent, that do not always match what the system actually does underneath.

Reddit confronted a related but distinct version of the same problem from the opposite direction this week. AdExchanger reported that brands have been generating what Bloomberg described as "stealth marketing content" across the platform, planting comments and posts about products in hopes the material gets picked up and repeated by Claude and ChatGPT as if it reflected genuine user opinion. Reddit's own disclosure, posted Monday, states that within one quarter its AI systems caught roughly 25,000 spammy posts and comments daily, contributing to a 20 percent year-over-year decrease in the low-quality content ordinary users actually see. MediaPost's own coverage of the same disclosure added that the same systems block 23 million spam views per day before reaching a human reader and have revoked nearly 2 million inauthentic votes over the preceding three months. Reddit attributed the increase in detected volume to more discerning tools rather than a rise in spam attempts themselves. The same MediaPost article cited a separate study released in May by AI communications firm 5W, which tracked more than 680 million AI citations across ChatGPT, Claude, Perplexity, Gemini and Google AI Overviews between August 2024 and April 2026, and found zero-click search rose from 56 percent of queries in 2024 to 69 percent by May 2025, a shift the study links to a 600 million monthly decline in traffic reaching news sites over the same period.

Shanzila Ahmed, chief business officer at generative engine optimization platform ReachLLM, told AdExchanger her agency has succeeded in developing Reddit posts for brand clients that were later surfaced by large language models, but that Reddit has been working actively to remove those same posts once identified. Shafqat Islam, president of AI marketing platform Optimizely, offered a sharper framing: paying people to mention a brand on Reddit, he told AdExchanger, violates what he called "the most sacred principles" of the platform, since Reddit's entire value to AI systems and to human readers alike rests on the assumption that its content reflects unfiltered, unpaid perspective rather than seeded promotion. The tension between those two positions, one from a practitioner who has built a business around achieving exactly this kind of placement and one from a rival platform executive characterizing the same activity as a violation of platform trust, captures the same underlying dynamic the FTC's accuracy statement is reaching toward from a different angle: AI systems increasingly function as a distribution channel that marketers want to influence, and the platforms supplying training and citation data to those systems are actively building defenses against exactly that influence.

Verification data shows fraud falling as targeting accuracy slips

A fourth thread running through the day's coverage concerns measurement rather than deal-making or regulation. DoubleVerify's publisher newsletter recirculated the company's Quarterly Benchmarks report for the first quarter of 2026 on July 7, PPC Land reported, and the headline figures move in genuinely different directions depending on which metric gets examined. Fraud and sophisticated invalid traffic violations fell 24 percent year-over-year to 0.5 percent globally, brand suitability violations declined 9 percent to 4.3 percent, and the global authentic viewable rate rose to 74 percent, up 3 percent from the same quarter a year earlier. Yet the rate of ads served outside their intended geography climbed 12 percent, reaching 1.2 percent globally, the only one of six headline quality metrics in the report that worsened rather than improved.

The report separates measurement into two distinct halves. Global Quality Benchmarks cover viewability, fraud and brand suitability, while a second section applying DoubleVerify's proprietary Attention Index measures how effectively advertisements hold a viewer's focus once technically visible on screen. Regional splits reveal considerable variation beneath the global averages: North America posted the strongest authentic viewable rate at 76 percent, followed by LATAM at 75 percent, EMEA at 68 percent, and APAC trailing the group at 63 percent. On the attention side, the ranking inverts. EMEA posted the strongest overall Attention Index among the four regions at 110, driven by an Engagement Index of 116, the highest engagement figure recorded across any region in the report, while North America posted the lowest regional Attention Index at 99, effectively matching the global baseline rather than exceeding it.

Device-level data illustrates why a single optimization rule cannot apply cleanly across both display and video formats. Mobile app environments produced the strongest display viewable rate of any device category at 82 percent, yet the same environment produced the lowest video viewable rate at 77 percent. A media buyer optimizing purely for viewability without accounting for format could end up allocating budget inconsistently across mobile inventory that looks similar on paper but performs in opposite directions depending on whether the creative is static or moving. Industry vertical rankings showed a similar pattern worth treating cautiously: Technology and Energy and Utilities tied for the strongest Attention Index at 107, while Telecom recorded the lowest score of any vertical at 90, a 17-point spread the report does not explain in terms of underlying causation.

The gap between an improving fraud rate and a worsening out-of-geo rate is worth sitting with on its own terms, because the two problems require entirely different fixes. Fraud detection depends on identifying non-human traffic patterns, while geo-targeting accuracy depends on correctly reading location signals and applying an advertiser's stated geography settings before an impression serves. A vendor or platform can improve steadily on the first problem while losing ground on the second, and this single quarter's data shows precisely that combination occurring at once.

DoubleVerify's benchmark release does not exist in isolation from the company's broader position in a verification market that has faced its own scrutiny. A shareholder derivative complaint filed in December 2025 alleges company executives misled investors about the effectiveness of its bot detection technology between November 2023 and February 2025, a legal proceeding entirely separate from the measurement methodology described in the quarterly report itself. A separate class action, filed earlier, made related allegations following a steep single-day stock decline after weaker-than-expected fourth-quarter results in early 2025. Those proceedings concern financial disclosures and investor communications rather than the measurement itself, and they sit apart from a format DoubleVerify has used consistently across prior benchmark releases.

The company's Fraud Lab has continued publishing technical investigations in parallel throughout this same stretch. A report published March 4, 2026 exposed a network the company calls AutoBait, documenting more than 200 domains using large language models and AI image generation to produce clickbait content at industrial scale and, in the estimate of three named DV Fraud Lab researchers who authored the disclosure, generating tens of millions of advertising impressions. The researchers included exposed operational code from the network itself, a level of technical detail that distinguished the report from earlier, less granular fraud investigations. Connected TV performance falls outside the scope of the Q1 2026 Quarterly Benchmarks document entirely, which explicitly excludes that channel from its video section, but DoubleVerify's separate 2026 Global Insights report, published May 7, found that CTV fraud schemes and variants rose 140 percent in the first quarter of 2026 compared with the same period a year earlier, and reported separately that more than one in three monitored CTV impressions were delivered to screens that were switched off entirely, a finding first detailed in company research published in March 2026.

The report's broader structure, splitting quality benchmarks from attention benchmarks into two separate halves, reflects a shift already underway in how advertisers evaluate media more generally. Viewability alone answers a narrow question: did an ad have the technical opportunity to be seen? Attention metrics attempt to answer a different one entirely: did anyone actually look at it, and for how long? That distinction carries practical consequences for campaign planning that extend well beyond this single report. A publisher or platform can post excellent viewability figures while producing mediocre attention scores, and the reverse holds equally true. The device-level data illustrates the point directly, since mobile app environments delivered the best display viewability of any device category in the report at 82 percent, yet the same environment produced the lowest video viewable rate at 77 percent. A media buyer optimizing purely for viewability, without reference to format, risks allocating budget inconsistently across otherwise similar-looking mobile inventory that performs in genuinely opposite directions depending on whether the creative moves or sits still.

The vertical rankings carry a comparable warning against oversimplified conclusions. A 17-point gap between the highest and lowest performing industries is a meaningful spread, but the report offers no explanation of causation. It does not say why Telecom, Travel and Automotive cluster at the bottom of the Attention Index, nor why Technology and Energy and Utilities cluster at the top. Advertisers in lower-scoring verticals may find the figures useful as a benchmark against which to measure their own campaigns, without treating the score itself as a diagnosis of what specifically needs to change. DoubleVerify's own methodology notes state the company indexes more than 100 billion impressions each month to build these global attention benchmarks, and all component indices are normalized to a baseline of 100 representing the average value across DoubleVerify's measurement ecosystem over a rolling 28-day window, meaning a score of 125 indicates performance 25 percent above that baseline while a score below 100 signals underperformance relative to the average.

Also noted

  • July 7, 2026: Google Search Console can now show how a site's social and video content performs within Google Search itself, adding Instagram, TikTok, X and YouTube content performance data to the platform property tooltip.
  • July 7, 2026: DoubleVerify's Attention Index framework, the same measurement system underpinning this week's quarterly figures, first extended into social platforms in June 2025 through a partnership with Snap, background PPC Land included alongside the new benchmark release.
  • July 7, 2026: Craig Graham, CEO of Grayvault Consulting, identified a new ChatGPT Ads feature letting advertisers upload custom audience lists under an "Audiences" tag, moving the platform closer to the first-party targeting already standard on Google and Meta, though Graham cautioned that match rates on uploaded lists remain partial rather than complete.
  • July 7, 2026: US Prime Day ad spend fell 8.8 percent while conversion rose 17.1 percent, CommerceIQ figures show, with shoppers browsing 10.3 percent fewer product pages yet buying more per visit and brands holding ROAS flat at 4.78x.
  • July 6, 2026: Sky confirmed it will acquire ITV's Media and Entertainment unit for 1.6 billion pounds, AdExchanger reports, a deal that would leave the combined entity accounting for 70 percent of the UK linear television advertising market once regulatory clearance arrives in the second half of 2027.