IAB Tech Lab chief executive Anthony Katsur has argued that the industry is telling the wrong story about Safari's domain block. In an essay titled "Apple Domain Blocking and the Cost to Publishers," posted on the Tech Lab's website on October 20, 2026 and also published on LinkedIn, Katsur wrote that the attention paid to the advertising technology companies caught by Apple's restrictions misses the larger effect on media companies that depend on advertising revenue. The essay asks Apple and other platform owners for clear policies, transition periods and review processes, points to Apple's own hiring for its advertising unit, and closes by presenting the Tech Lab's Trusted Server software as one response. It contains no figure for what publishers have lost, names no blocked company or domain, and cites no study.

In Short

The head of an advertising standards group wrote that when Apple's Safari browser blocks certain advertising companies, the websites that show those ads may earn less, and that this matters more than which tech vendor wins or loses. That affects news sites and other publishers that pay for their work with ads, because advertisers tend to move money to places where they can show an ad and prove it worked. The essay offers no numbers on how much money is at stake, so what it adds is an argument and a request to Apple, not a measurement.

What the essay says

The piece runs to about 1,300 words and is signed "Anthony Katsur, CEO, IAB Tech Lab." The Tech Lab's page carries the tag "trusted-server" and a note that "This article was also published on LinkedIn in October 2026." Page metadata on the Tech Lab site gives a publication time of 13:03 UTC on October 20. The essay appeared one day after Apple's WebKit team, in a post titled "Blocking Identity Resolution Infrastructure," said for the first time why Safari refuses to contact a set of advertising identity domains. Katsur does not mention that post by name.

His opening is a statement of priorities. "Apple's latest domain-blocking controversy has drawn attention to the advertising technology companies affected," Katsur wrote. "The bigger story is the pressure it puts on publishers worldwide. Ad tech can adapt, and I expect it will."

The second paragraph sharpens the point. "Too much of the industry conversation has treated this as a question of which technology vendor will prevail. That gets the business priorities wrong," according to Katsur. "The central concern should be media companies' ability to generate revenue and the public's continued access to the news, entertainment, services, research, and specialist information those companies provide."

Read against three weeks of reporting, the framing is notable for what it leaves aside. The Safari 27 block became public through the failure of The Trade Desk's ads to render, and the WebKit bug report that followed listed nine domains belonging to The Trade Desk, Unified ID 2.0, ID5, LiveRamp, Permutive and Audigent, as PPC Land detailed. Katsur's essay names none of them. It refers only to "the vendors named in the headlines."

"The technical scope continues to evolve"

Katsur's description of the mechanics is brief and deliberately general. "Recent reporting on Apple's restrictions describes disruption affecting advertising and identity services used across the open web," he wrote, with that sentence linked to an October 2026 Business Insider report on the WebKit update. "The technical scope continues to evolve."

That last line sits awkwardly beside the WebKit post of October 19, which defined the scope more tightly than anything Apple had said before. WebKit stated that Safari and WebKit in the iOS, iPadOS, macOS and visionOS 27 releases block network access to third parties that collect email addresses at login or newsletter sign-up and turn them into persistent tracking IDs, and that the blocking "is done on the domain level." The WebKit post named no company and published no list, and it said nothing about whether the list can be changed between software releases. So the scope is defined in purpose but not in membership. Katsur's essay addresses neither the definition nor the gaps. It does not use the phrase identity resolution, which is the activity WebKit says it is blocking.

For publishers, according to Katsur, the significance is plain: "restrictions that interrupt advertising demand or weaken measurement and attribution can make valuable audiences harder to monetize."

No figures on cost

The title promises a cost to publishers. The text does not quantify one. There is no estimate of lost revenue, no change in CPMs, no match rate or addressability figure for Safari traffic, no sample of publishers, no time period and no named publisher. The only quantitative references in the essay are implicit: "numerous openings" at Apple Ads and "weeks" for a technical remedy.

Instead Katsur sets out a chain of reasoning. The immediate risk, in his account, is lost demand. "If a buyer cannot reach an audience, measure advertising effectiveness, or worse, not be able to deliver an advertisement, they will take those ad dollars where they can do so," he wrote. "An opportunity to earn revenue exists when a reader visits. A technical remedy weeks later cannot recreate that moment or guarantee recovery of the income a publisher may have lost."

That passage reads as a reference to The Trade Desk's case, although the essay does not say so. Apple's John Wilander asked The Trade Desk to test an iOS 27.2 beta on October 5, 13 days after Apple first replied on the WebKit ticket, and the build appeared to restore the company's ad delivery domain. Identity domains were not reported as restored.

The longer-term risk, Katsur argued, is that publishers become harder to evaluate even when their ads still run. "As usable identifiers and exposure signals become scarcer, connecting a campaign to those results becomes harder," he wrote. "The publisher's audience may remain just as valuable. The evidence available to demonstrate that value may weaken."

Performance and brand measurement

The essay separates two buyer needs. "Performance advertisers and campaigns need to understand whether advertising contributed to purchases, subscriptions, or other actions. Brand advertisers and campaigns need to assess changes in awareness, recall, consideration, and intent," according to Katsur. Both, he wrote, benefit "from credible evidence about who was exposed and what followed."

He does not claim that measurement collapses without identifiers. "Meaningful measurement remains possible without a universal identifier," Katsur wrote. "Surveys, controlled experiments, aggregate analysis, and modeling all have a role. But alternatives can require more time, investment, or scale." His concern is distributional: "A smaller publisher may struggle to meet those requirements."

That concession matters, because it narrows the claim. The essay does not argue that a universal ID is necessary to sell advertising on Safari. It argues that losing one shifts the cost of proof onto publishers, and that the shift falls hardest on the smallest.

Where budgets go

The economic mechanism Katsur describes is one in which evidence, rather than performance, steers spending. "Buyers may discount inventory they find harder to measure or favor large platforms that can observe audiences and outcomes within their own environments," he wrote. "The risk is that spending follows the availability of evidence rather than the quality of the media or its actual contribution to a campaign."

It is a familiar argument in the industry, but the essay presents it as a risk rather than an observation. No data on budget movement since iOS 27 began rolling out on September 14 is offered, and neither Apple nor the Tech Lab has published any.

The final step in the chain is editorial. "Sustained pressure can force publishers to narrow coverage, postpone investment, increase subscription requirements, or cease operating," according to Katsur. He added a qualifier: "No single browser change explains these outcomes, but repeated monetization obstacles can make an already fragile business harder to sustain."

Privacy beyond the law

The essay's policy argument rests on the gap between platform rules and legal requirements. "Apple and other platform companies seek to establish consumer privacy protections that go beyond current government requirements," Katsur wrote. "Their policies may restrict practices that applicable law permits under specified conditions, including valid consent."

He acknowledged that there is no single benchmark. "Privacy law differs across jurisdictions, so there is no single global legal threshold against which every platform rule can be judged." The conclusion he draws is that "complying with the law and a consumer's expressed choice does not necessarily preserve access to the advertising capabilities a publisher needs."

The consent point echoes the position ID5's chief executive Mathieu Roche took when he called the block an attack on the web's business model and said regulators should look. It also runs into the WebKit Tracking Prevention Policy. As PPC Land noted in its coverage of the WebKit post, that policy treats certain visible logins as implied consent but says such logins "should require a user action and be noticeable by the user, not be invisible or hidden," and states that WebKit does not grant exceptions to specific parties. Apple's test, on that reading, is whether identification is noticeable to the user, not whether a consent record exists. Katsur's essay does not engage with that distinction.

"Companies are entitled to develop stronger privacy protections," he wrote. "But when those protections go beyond applicable legal requirements, they should also assess the economic consequences of their choices."

Apple's "Unintended Impact" section

Katsur draws directly on Apple's own policy text. "Apple's own WebKit policy recognizes this tension," he wrote. "It identifies website funding, advertising effectiveness measurement, and audience measurement among the activities that tracking prevention can unintentionally affect. That acknowledgment should carry practical weight."

That description is accurate as far as it goes. The policy's "Unintended Impact" section lists "Funding websites using targeted or personalized advertising", "Measuring the effectiveness of advertising" and "Audience measurement" among practices WebKit says it does not intend to disrupt. The same section also says: "When faced with a tradeoff, we will typically prioritize user benefits over preserving current website practices." Katsur's essay quotes neither line and does not mention the second. The policy's stated remedy is new browser technologies, with Private Click Measurement and the Storage Access API given as examples. Neither replaces cross-site identity.

Apple Ads hiring

The most pointed passage concerns Apple's own advertising business. "Apple's own investment in advertising adds another dimension," Katsur wrote. "Its careers site lists numerous openings across Apple Ads, including roles in machine learning, advertising relevance and quality, platform operations, and advertising sales." In the published version, that sentence links to a search of Apple's jobs site filtered to the Apple Ads team. The essay gives no count of openings and no date for the search.

He then drew a line under what the hiring shows. "These openings demonstrate that Apple is actively investing in its own advertising capabilities while setting browser rules that affect the wider market. Hiring does not establish that those rules were designed to disadvantage competitors."

What it does establish, in his telling, is a case for scrutiny. Hiring "does, however, reinforce the need for transparent standards and careful scrutiny of competitive effects when a company both governs access to audiences and sells advertising within its own services," according to Katsur. "Independent publishers should have a meaningful opportunity to compete for advertising budgets under rules whose privacy objectives and commercial effects can be assessed openly."

The essay is more restrained on this point than other critics. Raptive's Paul Bannister, a Tech Lab board member, linked the Safari block to Apple's 30% App Store fee and speculated about what Apple might charge ad companies next. Katsur stops short of any such claim.

Apple's advertising business does exist and is growing, by Apple's own qualitative account. Apple's Form 10-Q for the quarter ended June 27, 2026 attributed a 12% rise in services revenue, to $30.7 billion, "primarily" to advertising and cloud services, without breaking out advertising. Apple took Maps ads live on August 21. Every Apple Ads surface launched so far sits inside an Apple app, which means none of it is the publisher web inventory that the Safari block affects. Advertising Services Terms that took effect on July 28 did, however, extend the language on where Apple's ads may run to "web applications" and "other platforms or properties," without naming a new surface.

Apple's own public statements on the block come only from WebKit's published material, which presents it as a privacy measure and does not mention Apple's advertising business.

What Katsur asks of platforms

The requests are general. "Platform companies should work with publishers, advertisers, and technology providers before implementing changes with broad commercial consequences," Katsur wrote. "Clear policies, appropriate transition periods, practical review processes, and usable measurement alternatives can reduce unintended harm."

Four items, then: clear policies, transition periods, review processes and measurement alternatives. The essay sets no timeline for any of them, does not ask Apple to publish its domain list, and does not use the word appeal. It does not mention notification of listed companies.

That is a softer formulation than the one his colleague used a week earlier. Rowena Lam, the Tech Lab's senior director for privacy and data and the lead on Trusted Server, wrote on October 13 that the list "is not public facing," that listed domains "don't appear to be notified" and that "there does not appear to be any formal appeal process," as PPC Land reported. Lam also described an unreleased WebKit update that would let Apple add domains remotely. Katsur's essay does not repeat the remote-list claim, which neither Apple nor the WebKit code had confirmed when PPC Land reviewed it, and which the October 19 WebKit post did not address.

Katsur's closing test for any privacy measure is two-sided. "A privacy protection should be assessed for both its effectiveness and its impact on the legitimate businesses that finance the content consumers choose to access."

Trusted Server, again

The essay ends where much of the Tech Lab's public commentary on the block has ended. "That is why IAB Tech Lab built Trusted Server: to give publishers a tool for sustaining their advertising businesses as browser and mobile SDK restrictions evolve," Katsur wrote.

His plain-language description follows. "In plain language, it is open-source software that lets publishers handle more of the work of requesting, selecting, and delivering advertisements within systems they control, rather than relying as heavily on outside services running in a reader's browser or app. It also gives publishers greater control over how they share audience data with advertising partners."

The tone has shifted since early October. On October 8, Katsur posted on X that Trusted Server "is production ready TODAY!" and on October 9 that it moves "the entire open web ad stack server side" in "a turnkey solution," as PPC Land documented. The project's GitHub repository showed 343 commits and no published releases on October 10. The October 20 essay is more cautious. "Trusted Server is one tool in a publisher's tool belt. It does not guarantee revenue or resolve every restriction, and its use must respect consumer choices and applicable privacy requirements," according to Katsur.

"Selecting" advertisements

One word in the new description is worth noting. Katsur says Trusted Server lets publishers handle the work of "requesting, selecting, and delivering advertisements." In June 2025, the Tech Lab said that Trusted Server "does not conduct auctions" and complements Prebid Server. Whether "selecting" in the October 20 essay refers to running an auction, to choosing among responses from auction partners, or to something else is not explained. The essay gives no detail on architecture, deployment status, publisher adoption or results.

The circumvention question

The essay also says nothing about how a server-side setup interacts with WebKit's rules. The October 19 post states that a central resolver remains a third party "even if identity graphs are only scoped to data from a single company," and WebKit's policy says it may impose further restrictions "without prior notice" on parties that try to circumvent its protections. A browser cannot observe server-to-server traffic, so routing identity calls through a publisher's own domain avoids a third-party domain list. Whether Apple treats that as circumvention is a question the WebKit documents leave open and the Tech Lab essay does not raise.

Katsur frames the software's purpose in terms of agency rather than evasion: "Its purpose is to give publishers more agency when corporations take rulemaking into their own hands, extend beyond legal requirements, and impose restrictions that can fall unevenly across companies."

Who is speaking

IAB Tech Lab is a membership-based industry standards body whose members include advertising technology companies and publishers. Its board includes publisher executives such as Bannister. The essay carries the Tech Lab's own product tag, and its final section promotes software the Tech Lab maintains. Those facts do not make the argument wrong. They do mean the essay is advocacy from an interested party, not an assessment of the block's economic effect.

The essay also contains a critique of coverage, including this publication's. "Reporting on affected vendors is necessary, and some coverage has recognized the stakes for the open web," Katsur wrote. "The next step is to make publisher economics central to the story: how much demand remains available, how confidently advertisers can evaluate results, and whether smaller media businesses can bear the cost of adaptation."

Those three questions are measurable. Safari's share of publisher traffic is known to every site that runs analytics, and demand-side platforms and supply-side platforms see bid density and clearing prices by browser. None of that data appears in the essay, and the Tech Lab has not published any.

Why it matters for marketers

For advertisers and agencies, the essay's main practical claim is about measurement rather than delivery. If identity calls from Safari 27 return nothing, audience segments keyed to those IDs will not match on Safari traffic, and campaign reporting that depends on them will undercount reach. Katsur's point is that the cost of that undercount may be paid by the publisher whose inventory looks weaker, not by the identity vendor whose ID went missing.

For publishers, the essay confirms that the Tech Lab's response to Apple remains Trusted Server, now presented with fewer promises than three weeks ago. It also shows that the Tech Lab's formal request to Apple, at least in this essay, is for process: policies, transition periods, review and alternatives. It is not a request to reverse the block.

What it does not provide is the number that would test its own thesis. Without data on lost demand, changed CPMs or reduced match rates on Safari since September 14, the claim that publishers bear the larger cost remains an argument made by an industry body whose members include both sides of the trade.

Timeline

Summary

Who: Anthony Katsur, chief executive of IAB Tech Lab, the advertising industry's technical standards body, writing about Apple's Safari and WebKit domain block and its effect on publishers.

What: An essay arguing that the block's main cost falls on publishers through lost demand and weaker measurement, asking platforms for clear policies, transition periods, review processes and measurement alternatives, noting Apple Ads hiring while saying it does not prove intent, and presenting Trusted Server as one tool. It gives no cost figures, names no blocked vendor or domain, and cites no study.

When: Published on the IAB Tech Lab website on October 20, 2026, and on LinkedIn the same month, one day after WebKit's October 19 post on blocking identity resolution infrastructure.

Where: Safari and WebKit in Apple's iOS, iPadOS, macOS and visionOS 27 releases, affecting publishers worldwide that sell advertising to Safari users.

Why: The Tech Lab wants the debate over Apple's block to center on publisher revenue rather than on which ad tech vendor survives, and it is positioning its own open-source software as part of the response.