May 15 2026
GA4 Tracks AI Assistant Traffic, FAQ Results Gone – SEO Pulse via @sejournal, @MattGSouthern

Welcome to this week’s Pulse. The updates affect how you measure AI assistant traffic, what structured data does for visibility, and how a major publisher is planning for life after search.

Here’s what matters for you and your work.

Google Analytics Adds Native AI Assistant Channel

Google Analytics now assigns traffic from recognized AI chatbots to a dedicated “AI Assistant” default channel group. Custom channel groups with regex patterns are no longer the only way to separate AI assistant visits from referrals.

Key Facts

Sessions from recognized AI assistants now receive “ai-assistant” as the medium, route to a new “AI Assistant” default channel, and get a reserved “(ai-assistant)” campaign label. Google named ChatGPT, Gemini, and Claude as examples, but hasn’t published the full list of recognized referrers. All three changes happen automatically.

Why This Matters

Anyone running a custom channel group to isolate AI chatbot traffic can now compare their setup against Google’s native version. The custom regex patterns Google recommended last August still cover platforms outside the recognized referrer list. Both can run side by side.

The bigger question is what you do with the data once it’s visible. AI assistant traffic is now a distinct line item in acquisition, user, and channel reports. That makes it easier to compare conversion behavior, session quality, and volume against organic search without filtering or manual workarounds.

Google hasn’t said how quickly the recognized referrer list will expand as new platforms launch. If you track AI assistants beyond the three named examples, keep your custom groups running.

What Industry Professionals Are Saying

Kevin Indig, Growth Advisor at Growth Memo, commented on LinkedIn:

“Was about time! Literally complained about this on stage yesterday”

Johan Strand, Senior Digital Analyst and Partner at Ctrl Digital, wrote on LinkedIn:

“If you already have a Custom Channel Group set up to check for AI traffic, it´s probably a good idea to adapt it now.”

Read our full coverage: Google Analytics Adds AI Assistant As Default Channel Group

Google Completes FAQ Rich Results Deprecation

Google deprecated FAQ rich results, completing a removal that started a few years ago. The company added a notice to its FAQ structured data documentation without a blog post or separate explanation.

Key Facts

FAQ rich results stopped appearing in search results. Google will remove the FAQ search appearance filter in Search Console, the rich result report, and support for Rich Results Test in June. API support ends in August.

Why This Matters

If your reporting pipelines pull FAQ-specific data from the API, those API calls need to be updated before the August cutoff.

Leaving the markup in place shouldn’t create problems, but it no longer produces that visible result. Whether FAQ schema aids AI search is a separate question, and the deprecation doesn’t answer it.

Read our full coverage: Google Drops FAQ Rich Results From Search

Ahrefs Report: Adding Schema Didn’t Increase AI Citations

An Ahrefs report tracked 1,885 pages that added JSON-LD schema and found no meaningful increase in AI citations across Google AI Overviews, AI Mode, or ChatGPT.

Key Facts

Ahrefs matched each treated page against controls that never added schema and measured changes over 30-day windows. AI Overviews showed a 4.6% decline relative to controls, while AI Mode (+2.4%) and ChatGPT (+2.2%) showed changes too small to distinguish from noise.

Why This Matters

The correlation between schema and AI citations has been widely cited as evidence that structured data improves AI visibility. Ahrefs tested whether the relationship appeared causal and found no evidence of a meaningful lift, at least for pages already being cited. Sites with schema tend to also invest in better content, stronger authority, and more links. Those factors may explain the correlation better than the markup itself.

The report can’t say whether schema helps pages that aren’t yet visible to AI systems. That’s a different population that needs its own test. For pages already earning citations, though, adding JSON-LD is unlikely to be the unlock.

What SEO Professionals Are Saying

Chris Long, Co-founder of Nectiv, wrote on LinkedIn:

“this data is changing my viewpoint a bit on how effective it is at actually influencing AI citations.”

Read our full coverage: Schema Markup Didn’t Move AI Citations In Ahrefs Test

Condé Nast CEO: Plan As If Search Traffic Will Be Zero

Condé Nast CEO Roger Lynch said he told company teams to plan their businesses as if search traffic were zero. Lynch made the comments in an interview on TBPN, a tech talk show OpenAI acquired in April.

Key Facts

Lynch described three consecutive years in which internal forecasts underestimated the actual declines in search traffic. He expects search to settle at a single-digit percentage of total traffic, not literally zero.

Lynch pointed to a “barbell effect” in which large, authoritative brands and small, niche publications are performing well, while brands in the middle are most exposed. Condé Nast’s digital subscriptions grew 29% in revenue last year.

Why This Matters

Lynch is describing what the third-party data has been showing for months. Chartbeat reported a 60% decline in search referrals for small publishers over two years. The Reuters Institute found that media leaders expect search traffic to fall by more than 40% over three years. The difference is that a CEO running Vogue, The New Yorker, and GQ is now building budgets around those numbers.

The barbell observation is worth testing against your own client portfolio or publishing operation. Lynch’s argument is that brands without deep category authority or a strong niche focus lack a clear path forward. AI Overviews, commerce links, and sponsored results fill the page before organic listings appear.

What SEO Professionals Are Saying

Kevin Indig, Growth Advisor at Growth Memo, commented on LinkedIn:

“Makes sense, no escape hatch for publishers in AEO.”

Read our full coverage: Condé Nast CEO: Plan As If Search Traffic Will Be Zero

Theme Of The Week: The Measurement Is Catching Up To The Problem

The tools and signals that defined search visibility for years are being deprecated, questioned, or abandoned by the publishers who depended on them.

FAQ rich results are gone. Schema’s role in AI citations is weaker than the correlation suggested. A major publisher is planning as if search traffic won’t recover. Each story involves an environment where the old measurement infrastructure no longer matches the landscape.

The GA4 update is the other side of that coin. Google is building native tracking for the traffic source that’s growing while the traditional one contracts.

AI assistant traffic is a fraction of what search delivers. But it’s now visible by default, in the same reports, next to the channels it’s measured against.

Top Stories Of The Week:

More Resources:

Featured Image: PeopleImages/Shutterstock; Paulo Bobita/Search Engine Journal

Ecommerce MGMT 0 Comments
May 14 2026
Google Analytics Adds AI Assistant As Default Channel Group via @sejournal, @MattGSouthern

Google Analytics added an “AI Assistant” default channel group for traffic from recognized AI chatbot referrers, with Google naming ChatGPT, Gemini, and Claude as examples.

GA4 property owners no longer need to build custom channel groups with regex patterns to separate AI assistant visits from referrals. Until now, all AI chatbot traffic landed in the Referral bucket by default.

What’s New

The update touches three traffic source dimensions at once.

When Google Analytics detects a referrer matching a recognized AI assistant, it assigns “ai-assistant” as the medium value. Those sessions then get grouped under the “AI Assistant” channel in Default Channel Group reports. The campaign dimension receives a reserved “(ai-assistant)” label.

All three changes happen automatically. Property owners don’t need to configure anything.

Google described the update as a way to “monitor how generative AI impacts your business by tracking user clicks, trending AI sources, and how this traffic compares to traditional channels like organic search.”

Google hasn’t published the full list of recognized AI assistant referrers. The Help Center entry names ChatGPT, Gemini, and Claude as examples.

Context

Google has been working toward this for almost a year. In August, the Analytics team published guidance on building custom channel groups with regex patterns to capture AI assistant traffic. That guidance named ChatGPT, Gemini, Microsoft Copilot, Claude, and Perplexity as platforms to track. That marked the point when Google’s own documentation began treating AI assistant traffic as a category worth measuring separately.

The custom channel group workaround had limitations. Regex patterns required manual maintenance as AI platforms changed domains. Property owners needed editor-level access to set them up. And the two-custom-channel-group limit in GA4 meant dedicating one of only two available slots to AI tracking.

This follows a pattern Google set in 2022 when it added “cross-network” as a default channel group to capture Performance Max and Smart Shopping traffic. That update also moved traffic out of a generic bucket into its own dedicated channel without requiring manual configuration.

AI traffic attribution has been a recurring measurement challenge. Last year, Google fixed a bug that caused AI Mode search traffic to be reported as “direct” instead of “organic” in GA4 after a noreferrer code was stripping referrer headers. Google also added AI Mode data to Search Console performance reports, though that traffic gets blended into existing totals rather than appearing as a separate category.

Why This Matters

Anyone running a custom channel group to track AI assistant traffic may be able to simplify that setup as the native channel appears in reports. The native channel may reduce the need for the regex patterns and manual channel ordering that Google recommended last year.

Properties without custom AI tracking will start seeing this traffic broken out from referrals automatically. Sessions that previously appeared as generic referral traffic from chatgpt.com or claude.ai will have their own channel.

One gap worth watching is the referrer limitation. AI assistant traffic that arrives without a referrer header still lands in Direct. This can happen through in-app browsers and mobile apps, or when users copy and paste links. The new channel only captures what GA4 can identify through the referrer.

Looking Ahead

Google hasn’t published which AI assistants are on the recognized referrer list beyond the three named examples. It also hasn’t said how the list will be updated as new platforms launch. The August 2025 custom channel group guidance named five platforms, but the new automatic system doesn’t specify its full coverage.

The Default Channel Group definitions page hasn’t been updated to include “AI Assistant” in its channel table yet, so the full technical definition isn’t available to review. The custom channel group regex patterns Google published last year can still cover platforms that aren’t on the recognized referrer list.

Featured Image: Stocking/Shutterstock

Ecommerce MGMT 0 Comments
May 13 2026
Liquid Web WordPress Plugin Rebrand Triggers Backlash via @sejournal, @martinibuster

Liquid Web inadvertently started a cascading series of controversies after it folded a group of well-known WordPress plugin brands into a new software lineup. The reconfiguration and rebranding caught users by surprise, leading to confusion and significant online backlash against Liquid Web across social media.

A Dynamic WordPress Facebook group admin started a discussion about the Liquid Web plugin and branding controversy that reflected the confusion at the time, writing:

“It looks like a bit of chaos in the Kadence FB group as LiquidWeb moves to integrate their tools under a single umbrella. What’s interesting is that they’ve dropped Lifetime Bundle (LTD) and now have 3 packages:

  • $99 Essentials (theme and blocks),
  • $219 Pro (which includes ShopKit)
  • and $399 Elite.

Ultimately, people aren’t happy. It appears that their licenses aren’t working. That’s something they should be able to fix. However, it’ll be interesting to see what level of access people get. Will LTD owners still retain access to addons like ShopKit and Kadence Conversions?”

One person in that Dynamic WordPress Facebook group discussion blamed private equity investments in web hosting for the issue, a sentiment that was echoed on X, where @jeffr0 suggested that maybe Matt Mullenweg had a point about private equity firms and WordPress hosting investments.

@jeffr0 tweeted:

“So I guess @photomatt had a point. Private Equity in the WordPress ecosystem blows.”

Someone else disagreed with blaming private equity investors, responding:

“I’m not sure I agree. First, WPE wasn’t doing anything wrong. …I’m also not sure what’s happening to these plugins is a result of LW being owned by PE.”

Reflecting the confusion in the moment, @srikat tweeted:

“I can’t find the downloads for my lifetime Kadence purchase. Just sent them a support email ticket..”

Nexcess/Liquid Web Branding and Rebranding

Part of the confusion stems from a yearslong series of Liquid Web and Nexcess branding flip-flops.

  • Liquid Web acquired Nexcess in 2019.
  • The two brands later moved toward a unified Liquid Web identity.
  • By late 2025, users who typed in nexcess.net were often redirected to liquidweb.com.
  • In April 2026, Nexcess relaunched as a “Specialty Cloud” brand combining Liquid Web’s managed hosting expertise with Servers.com’s bare-metal infrastructure.

Meanwhile, Liquid Web is now the managed hosting brand within the Nexcess ecosystem.
StellarWP Disappears: Plugins Emerge In Nexcess And Liquid Web
Previously, the plugins lived under the StellarWP brand, with many maintaining their own standalone websites. The new branding is confusing for some users because both Nexcess and Liquid Web describe the same WordPress products as part of their own ecosystem.

The Nexcess relaunch announcement from April 8, 2026 says:

“We’re expanding your toolkit by bringing leading software solutions, like Kadence, GiveWP, The Events Calendar, and LearnDash, directly into the Nexcess ecosystem.”

Liquid Web’s May 12, 2026 web page describes the same products as part of the Liquid Web by Nexcess software portfolio:

“Liquid Web by Nexcess is concentrating its diverse WordPress software portfolio into four core products…”

The overlapping language between both brands helps explain why the rollout appeared confusing from the outside. The products were described as moving into both Nexcess and “Liquid Web by Nexcess,” and StellarWP seemingly disappeared without notice.

Liquid Web’s software announcement says its WordPress software portfolio is now concentrated into four core products: Kadence, LearnDash, The Events Calendar, and Give. The company says SolidWP, Iconic, Restrict Content Pro, and MemberDash are no longer sold as standalone products, with their features folded into Kadence or LearnDash.

What It Means For Plugin Subscribers

For existing customers, Liquid Web says the change is optional. The company says customers can keep their current features, plans, pricing, tools, and license keys unless they choose to upgrade to one of the new software plans.

But the public rollout appears to have created confusion among plugin users, including lifetime deal customers who were unsure what happened to the products they had purchased. Social media posts described product pages disappearing, redirects not working as expected, and users trying to determine whether their plugins had been discontinued, renamed, or moved.

In a post to a discussion in the Dynamic WordPress Facebook group, Jack Kitterhing, Strategic Product Leader at Nexcess, confirmed that lifetime subscription plugin customers would retain what they already had and that every customer was being grandfathered in. He also acknowledged login issues and missing invoices, describing the move as a “massive migration and change of systems” that came with challenges.

Kitterhing posted an explanation of what’s going on:

“Just to confirm Lifetime customers retain everything they already had. We aren’t removing anything or watering it down. If you owned it you still own it today. Every single customer is being grandfathered in.

And we re-positioned Kadence Essentials so for those of you who just want the theme and blocks it’s now cheaper than it used to be ($99 vs $129) to get the core components of Kadence.

There are currently issues with logging in for some customers and missing invoices which the team is fixing as I type and we expect to be fully fixed in a few hours.
This was a massive migration and change of systems and like anything of such magnitude it comes with challenges. Thanks for bearing with us as we get this all up and running today.”

Takeaways

  • Liquid Web says existing customers keep their current features, pricing, plans, tools, and license keys.
  • Lifetime customers were told they retain what they already had.
  • The backlash appears to have been driven by confusion during the rollout, not only by the product consolidation itself.
  • Years of Liquid Web and Nexcess branding changes made the plugin migration harder to understand.
  • Clearer advance communication may have reduced the confusion around product pages, redirects, licenses, and lifetime deal access.

There appears to have been insufficient communication from Liquid Web and Nexcess, compounded by the two companies’ branding flip-flops. The situation appears to be on the way to being resolved.

Featured Image by Shutterstock/hoangpts

Ecommerce MGMT 0 Comments
May 13 2026
Google Quietly Changed How Search Terms Are Reported For Some AI Queries via @sejournal, @brookeosmundson

Google quietly updated one of its Google Ads help pages with a clarification that could raise concerns for some advertisers.

The updated documentation suggests that search terms shown in reporting for AI-powered Search experiences may not always reflect a user’s exact query. Instead, some reported search terms may represent Google’s interpretation of user intent.

The change applies to experiences tied to AI Mode, AI Overviews, Google Lens, and autocomplete.

Search Terms Reports have long been used to understand query intent, identify negative keywords, review compliance concerns, and spot optimization opportunities. While the report has never provided full visibility, advertisers generally assumed that when a search term appeared in reporting, it reflected the actual query entered by the user.

For some newer AI-powered Search experiences, that may no longer be the case.

What Google Changed

The updated language appears within Google’s help documentation around ad group prioritization. The page explains how Google determines which ad group enters an auction when multiple keywords or targeting methods are eligible to match the same search.

It was first discovered by Anthony Higman who posted about his findings on LinkedIn.

Within that documentation, Google now explains that search terms associated with AI-powered experiences may reflect the inferred meaning or intent behind a search instead of the literal query itself. The clarification specifically references AI Mode, AI Overviews, Lens, and autocomplete.

In practice, that means advertisers could see search terms in reporting that were never directly typed by the user. Instead, Google may surface a normalized or interpreted version of the interaction.

Historically, many advertisers viewed the Search Terms Report as a fairly direct reflection of user behavior. A user searched for something, a keyword matched, and the advertiser could review that query inside reporting.

For some AI-powered Search experiences, Google is now signaling that the reporting process may involve more interpretation before those search terms appear in the interface.

Why Google Likely Made This Change

This update likely reflects the practical challenges of reporting on newer AI-powered Search experiences, especially with the recent announcements of more ads coming to AI experiences.

Traditional Search reporting was built around direct keyword queries. AI-powered experiences like AI Mode, AI Overviews, Lens, and autocomplete do not always work that way.

Users may refine searches across multiple prompts, search visually instead of typing, or rely on autocomplete suggestions before finishing a query. In some cases, there may not be a single clean keyword query for Google to surface inside a traditional Search Terms Report.

From Google’s perspective, intent approximations may help standardize reporting across those interactions. A conversational AI search, a Lens query, and an autocomplete-assisted search may all require some level of interpretation before they can appear in reporting.

There’s probably also a privacy component to this.

As Search becomes more conversational, users naturally provide more context in their interactions. Google may not want to expose every raw AI prompt, image-based search, or conversational refinement directly inside advertiser reports.

Many advertisers will likely understand that reasoning. The problem is that some may also see this as another reduction in transparency at a time when Google Ads already relies heavily on automation, modeling, and inferred signals.

Should Advertisers Be Concerned About This Change?

Many advertisers will likely view this as part of a broader trend inside Google Ads.

Over the past several years, advertisers have already adjusted to reduced search term visibility, heavier automation, broader matching behavior, and more modeled reporting. This update adds another layer to that shift by signaling that some visible search terms may not represent the exact user query.

For advertisers who rely heavily on search term analysis, that creates obvious concerns.

Highly regulated industries often review search terms closely for compliance and brand safety. B2B advertisers use query reports to identify customer pain points and emerging use cases. Ecommerce advertisers use Search Terms Reports to build negative keyword lists, refine product segmentation, and better understand shopping behavior.

If reported terms become interpreted summaries instead of direct queries, advertisers may start questioning how confidently they can optimize against that data.

There are also still several unanswered questions around how these approximations actually work.

Google has not publicly explained how much interpretation occurs, whether advertisers can distinguish modeled terms from literal queries, how negative keywords interact with interpreted intent, how closely approximated terms reflect the original user phrasing, or whether reporting consistency could change as AI models evolve.

That lack of detail will likely make some advertisers uneasy.

A marketer could review a search term report and assume they are looking at direct customer language when the term may actually represent Google’s interpretation of the interaction. That distinction matters when advertisers are making optimization decisions, reviewing compliance concerns, or reporting insights internally.

Some Advertisers May Be Comfortable With This Change

On the other hand, there’s probably lots of advertisers who won’t see this as a big deal.

Some advertisers already optimize more around intent themes, conversion quality, and broader performance patterns than exact query language. For accounts heavily using broad match and Smart Bidding, interpreted search terms may not feel dramatically different from how optimization already works today.

There is also a practical challenge Google is trying to solve.

AI-powered Search interactions do not always produce simple keyword queries that fit neatly into traditional reporting. In some cases, a normalized intent summary may actually be easier for advertisers to review than fragmented conversational prompts or image-based searches.

That does not remove the transparency concerns, but it does help explain why Google may view interpreted reporting as a necessary adjustment for AI-powered Search experiences.

What Does This Mean For Future Optimization?

This update may push advertisers to rely less on literal query analysis over time, especially as more Search activity moves into AI-powered experiences.

For years, Search optimization has centered heavily around search term analysis. Advertisers mined queries for negatives, refined match types, identified customer language, and built campaign structures around tightly grouped intent.

If Search Terms Reports increasingly include interpreted intent instead of direct queries, some of those workflows may become less precise.

Optimization may shift further toward broader signals like landing page alignment, first-party data, conversion quality, audience behavior, CRM integrations, and overall content relevance.

That doesn’t make search term reports useless, though.

Advertisers may need to treat them more as directional insight rather than exact representations of customer language.

This could also change how marketers communicate reporting internally.

Many teams still use Search Terms Reports to demonstrate customer intent to executives, clients, or other stakeholders. If some reported terms now reflect modeled interpretations instead of literal searches, marketers may need to be more careful about how those insights are presented and explained.

A reported term may still reflect the general intent behind a search. It just may not represent the exact words the customer used.

Looking Ahead

This documentation update may end up being more important than it initially appears.

Search Terms Reports have long been one of the few places advertisers could directly connect user queries to campaign behavior. Google is now signaling that some of those reported terms may involve interpretation before they appear in reporting.

That will likely become more noticeable as AI-powered Search experiences continue expanding across Google Search.

For advertisers, the bigger issue may simply come down to clarity. If interpreted search terms become more common, many advertisers will likely want more visibility into how those terms are generated and how closely they reflect actual user behavior.

Featured Image: vittaya pinpan / Shutterstock

Ecommerce MGMT 0 Comments
May 13 2026
Condé Nast CEO: Plan As If Search Traffic Will Be Zero via @sejournal, @MattGSouthern

Condé Nast CEO Roger Lynch says he told company teams to plan their businesses as if search traffic were zero.

Lynch made the comments in an interview on TBPN, a tech talk show OpenAI acquired in April. He described three consecutive years in which internal budget forecasts underestimated actual declines in search traffic.

Lynch said:

“Each of the last three years, we would do our budgets, and we’d put forecasts in of search traffic declining… Because we’d seen the pattern of algorithm changes. And generally those algorithm changes were negative.”

“Every year, our search traffic was down more than we had forecast. So last year I told our teams, ‘Assume there’s no search.’ You have to have your businesses planned as if search is zero.”

Lynch told TBPN that Condé Nast doesn’t expect search traffic to literally reach zero. He expects it to settle at a single-digit percentage of total traffic.

What Changed

Lynch described how the search results page has changed, based on a comparison his team prepared for a recent board meeting. Lynch recalled:

“We took a snapshot of search results from seven or eight years ago. And what you saw were a few sponsored links, then the ten blue links.”

“Do the same search today, you get an AI overview, then you get rows and rows and rows of commerce links, then you get sponsored stuff.”

He noted that someone had recently asked him how search revenue could be up. “Have you done a search recently?” Lynch replied. “I basically have to go to the second page to get an organic result.”

Lynch acknowledged that changes in search traffic have affected Condé Nast’s business. The company has continued to grow revenue and profitability despite the decline, which he called a “headwind” rather than a crisis.

The Barbell Effect

Lynch described what he called a barbell effect across the Condé Nast portfolio. In his telling, large, authoritative brands and small niche publications with loyal audiences are performing well. Brands caught in the middle are the most exposed.

“Vogue has grown every year I’ve been at the company. It grows revenue, grows profitability every year,” Lynch said.

The New Yorker had its most successful year ever, he added. On the other end, Lynch pointed to Pitchfork, which represents about 1% of Condé Nast’s revenue but has a loyal audience in its category.

Lynch explained:

“If you try to be too broad, too large of an audience, this is not the era for that… You either need to be large and authoritative in a big category… or you need to be really nailing a specific niche where you have a loyal audience that’s willing to pay.”

Lynch added that brands in the middle of that barbell, those without deep authority in a category or strong enough niche focus, don’t have a clear path forward.

He added:

“If you don’t have really strong authoritative brands, or brands that have very strong niche in certain areas, or direct audiences, then you’re just going to be fighting that all the way down.”

Subscriptions As The Replacement

Condé Nast’s digital subscriptions grew 29% in revenue last year, according to Lynch. The company reported double-digit growth, which is continuing this year.

Lynch noted the company has raised subscription prices “fairly materially” over the past couple of years. He expected retention to decline with each increase. Instead, retention improved every year.

The company is also expanding subscriptions to smaller brands. Pitchfork and Tatler both launched paid digital subscriptions recently.

Why This Matters

Lynch’s comments are consistent with third-party measurements indicating that publisher search referrals are under pressure. Chartbeat data reported in March showed search referral traffic fell 60% for small publishers over two years. A Reuters Institute survey found media leaders expect search traffic to decline by more than 40% over three years.

Google’s VP of Search, Liz Reid, has reframed those losses as reductions in low-quality “bounce clicks.” Google hasn’t shared publisher-facing data to support that claim.

Lynch’s directive carries weight because of the portfolio behind it. Condé Nast operates Vogue, The New Yorker, GQ, Vanity Fair, Architectural Digest, Condé Nast Traveler, Wired, and Pitchfork, among others. When the CEO of a portfolio that includes those brands says teams should budget for zero search traffic, it gives industry data a concrete example from a major publisher.

The barbell observation matters for anyone managing a publisher caught between the two extremes. Lynch is describing a version of the pressure Chartbeat’s size-segmented data has tracked. Small and mid-tier publishers without deep category authority or direct audience relationships face the steepest declines.

Looking Ahead

Lynch told TBPN the company has started evaluating each brand’s plan for a low-search future. The company is prioritizing brands that can show a path forward without search traffic.

Lynch’s comments may put pressure on other large publishers to formalize similar planning. The trend data has been consistent enough that budgeting for search decline is already common. Budgeting for zero is a different level of preparation.

Ecommerce MGMT 0 Comments
May 13 2026
Google Research’s ALDRIFT: AI Answers That Do More Than Sound Plausible via @sejournal, @martinibuster

Google Research published a paper that studies how to make generative AI systems produce answers that do more than sound plausible. The researchers say that their ALDRIFT framework “opens exciting avenues” for moving beyond answers that merely have a high probability.

The paper, titled “Sample-Efficient Optimization over Generative Priors via Coarse Learnability,” examines a problem in which generated answers must remain likely under a model while also moving toward a separate goal. The research points toward new avenues for addressing the AI plausibility trap.

Google ALDRIFT

The evidence in the paper centers on a framework called ALDRIFT (Algorithm Driven Iterated Fitting of Targets). The method repeatedly refines a generative model toward lower-cost answers and uses a correction step to reduce accumulated error during the process.

The paper also introduces “coarse learnability.” The term means the learned model does not need to perfectly match the ideal target. It needs to keep enough coverage over important parts of the answer space so useful possibilities are not lost too early. Under that assumption, the authors prove that ALDRIFT can approximate the target distribution with a polynomial number of samples.

ALDRIFT Operates On A Two-Part Setup

ALDRIFT operates on a two-part setup:

  1. The generative model represents what kinds of answers remain likely under the model.
  2. The outside scoring process measures whether a candidate answer performs well against the target goal.

The authors describe that score as a “cost.” The word “cost” refers to the measured penalty assigned to a candidate answer. A lower cost means the candidate did better according to the requirement being checked. ALDRIFT does not simply search for any low-cost answer. It searches for answers that score well while still remaining likely under the generative model.

Some AI Answers Need To Work As A Whole

The researchers are focused on AI answers for problems where the response has to function in the real world such as their examples of route planning and conference planning.

  • Route planning: The paper explains that an LLM may evaluate whether individual route segments are scenic, but may struggle to ensure that those segments connect into a valid path.
  • Conference planning: An LLM may group sessions by topic, while a classical algorithm may be needed to schedule those sessions into a timetable without conflicts.

These examples show why the paper treats plausible answers as only part of the problem. The harder issue is producing answers that remain coherent when separate parts have to work together as one complete solution.

The Coarse Learnability Assumption

The paper treats this as a problem of guiding a generative model toward answers that hold together across all their parts. The authors connect the problem to inference-time alignment, where a model is adjusted during use based on whether a specific answer works as a complete solution. That connection gives the research practical relevance, although the paper’s contribution remains theoretical and depends on the coarse learnability assumption.

The phrase “coarse learnability assumption” means the paper’s theory depends on an assumption that the model can keep enough useful possibilities available while it is being pushed toward better answers.

It does not mean the model has to learn the target perfectly. It means the model has to preserve enough coverage of the answer space so the process does not get stuck too early or lose possible better answers.

Existing Optimization Methods Leave Sample-Limited Gaps

The paper identifies several gaps in how existing optimization methods are understood:

  • Limitation of existing methods: Classical model-based optimization methods rely on “asymptotic convergence arguments.” This means they are theoretically understood after very large amounts of sampling, but not necessarily in practical settings with limited samples.
  • Failure with expressive models: The paper says these classical assumptions “break down” when using expressive generative models like neural networks.
  • Gap in understanding: The authors say the “finite-sample behavior” of optimization in this setting is “theoretically uncharacterized.” That means the theory does not fully explain how these methods behave when only limited samples are available.

The paper’s solution is to introduce “coarse learnability” to explain how a generative model can be pushed toward better answers while keeping enough useful possibilities available along the way.

The LLM Evidence Is Limited

The paper’s main proof applies to analytic generative models, which are easier to analyze mathematically than modern LLMs. The LLM evidence is narrower: the authors use GPT-2 in simple scheduling and graph-related problems, showing behavior that supports the idea without proving that the same assumptions hold for modern LLMs.

The Research Points To A Foundation For Future Research

The paper offers a theoretical foundation for studying how generative models could be combined with external checking processes.

The research shows that Google researchers are exploring a framework for addressing the “plausible answer” problem, and the authors write that the “framework opens exciting avenues for future research.” They conclude that this research points “toward a principled foundation for adaptive generative models.”

Takeaways

  • The “Coverage” Requirement:
    Coarse learnability means the model does not have to learn the target perfectly. It needs to avoid losing useful areas of the answer space where better solutions might exist.
  • The Correction Step Matters:
    ALDRIFT uses a correction step to keep the search closer to the intended target as the model is pushed toward better answers.
  • Two-Part Approach:
    The framework uses a division of labor. The generative model handles qualitative or semantic preferences, while a separate process checks whether the answer works as a complete solution.
  • Limited LLM Evidence:
    Tests with GPT-2 showed behavior that supports the idea in simple scheduling and graph-related examples, but not proof that the same assumptions hold for modern LLMs.
  • Real-World Use Is The Larger Goal:
    The research matters to SEOs and businesses because AI answers are increasingly expected to do more than summarize information. They need to support decisions, plans, and actions that hold together outside the chat interface. While the framework is likely not being used in production, it does show Google is making progress on providing answers that are more than plausible.

Read the research paper here:

Sample-Efficient Optimization over Generative Priors via Coarse Learnability (PDF)

Featured Image by Shutterstock/Faizal Ramli

Ecommerce MGMT 0 Comments
May 11 2026
Schema Markup Didn’t Move AI Citations In Ahrefs Test via @sejournal, @MattGSouthern

Schema markup is far more common on pages cited by AI. But a new Ahrefs report found that adding it didn’t result in a clear increase in citations.

Ahrefs tracked 1,885 web pages that added JSON-LD schema. Each page was matched against control pages that never added schema, and citation changes were measured across Google AI Overviews, AI Mode, and ChatGPT.

No platform showed a meaningful citation increase after schema was added.

What Ahrefs Found

The report analyzed 6 million URLs and found that pages cited by AI were roughly three times more likely to include JSON-LD. This gap has been seen as evidence that schema improves AI visibility. However, Ahrefs tested whether this held true when isolated from other signals, since sites with schema tend to invest in better content and earn more links.

They ran a controlled comparison, matching each schema page with three control pages from different domains with similar citation levels that never added JSON-LD. Citation changes were measured 30 days before and after schema addition.

Using its Brand Radar tool and Agent A, Ahrefs conducted a matched difference-in-differences analysis to account for platform trends. Here’s what was found.

  • Google AI Overviews: −4.6% (a small but statistically notable decline relative to controls)
  • Google AI Mode: +2.4% (too small to distinguish from random variation)
  • ChatGPT: +2.2% (too small to distinguish from random variation)

Three more tests were run alongside the primary comparison, and all four found no clear positive or negative effect.

The AI Overview Decline

The −4.6% decline in the AI Overview section deserves context. Ahrefs reports both treated and control pages were already declining before schema was added. Treated pages declined slightly faster, but the difference is small, with about 12 fewer daily citations per page in a sample where most pages received hundreds.

The report notes that the decline could reflect a small negative effect from schema, or it could be coincidence. It doesn’t draw a conclusion either way.

What The Report Doesn’t Cover

Every page in the dataset had 100+ AI Overview citations before any schema was added. These pages were already in the consideration set, being crawled and surfaced.

The report admits this limitation. For pages not yet visible to AI, schema might still aid crawling, parsing, or indexing, but the data can’t confirm this.

The report also notes other limitations. Pages adding JSON-LD often change other elements, making it hard to separate schema effects from those changes. All schema types were pooled, so some might perform differently. The 30-day window might miss slower effects.

A searchVIU experiment cited in the report tested whether five AI systems used schema markup when fetching pages in real time. None did; they only extracted visible HTML, ignoring JSON-LD, Microdata, and RDFa. This was a direct-fetch test, not proof of schema’s role during training, indexing, or retrieval.

Why This Matters

Schema markup is frequently recommended for AI visibility. However, Ahrefs’ data complicates this. While schema supports rich results and knowledge graphs, adding JSON-LD doesn’t increase AI citations for pages already cited.

The data shows a correlation: pages with schema are cited more often by AI, but Ahrefs interprets this as a sign of overall site quality rather than schema’s direct impact.

Looking Ahead

The report can’t determine whether schema helps pages that aren’t yet cited, which is a different group of pages that need another study. If pages are visible to AI, JSON-LD probably won’t boost citations.

Featured Image: Roman Samborskyi/Shutterstock

Ecommerce MGMT 0 Comments
May 10 2026
Google Drops FAQ Rich Results From Search via @sejournal, @MattGSouthern
  • Google has removed FAQ rich results from Search after limiting them for most sites.
  • Search Console reporting ends in June, followed by API support in August.
  • FAQ schema can stay on your pages, but it no longer earns visible FAQ results in Google.

Google deprecated FAQ rich results, completing a removal that started years ago. FAQ rich results were already restricted for most sites.

Ecommerce MGMT 0 Comments
May 8 2026
New AI Search Links, Core Update Winners And Losers – SEO Pulse via @sejournal, @MattGSouthern

Welcome to the week’s Pulse: updates affect how links appear inside AI search results, which types of sites gained or lost visibility in the March core update, and how Google’s Preferred Sources feature works across new markets.

Here’s what matters for you and your work.

Google Adds More Links And Link Context To AI Search

Google announced five updates to how links appear in AI Overviews and AI Mode.

Key facts: The updates add more inline links, more links at the end of some AI responses, previews from public forum discussions, and desktop hover previews.

Why This Matters

Inline links placed next to the text they support could change the click math for pages cited in AI results. Until now, most AI Overview citations have clustered at the bottom of the response, where they compete with one another and are easy to skip. Placing them closer to the relevant sentence gives each link more context.

Previews from public discussions add a new surface for content from Reddit, forums, and similar platforms. If your brand or product is discussed on those platforms, that content may now appear alongside AI-generated answers with your name attached.

Read our full coverage: Google Adds More Links & Link Context To AI Search

Core Update Data Shows Aggregators Losing Ground

An analysis from Amsive found that aggregators and user-generated content platforms lost US search visibility after Google’s recent core update, while first-party brand sites and government domains gained. Lily Ray, VP of SEO and AI Search at Amsive, examined over 2,000 domains using SISTRIX Visibility Index data.

Key facts: YouTube lost 567 SISTRIX visibility points, the largest single-domain drop in the dataset, and roughly 30% larger than Wikipedia’s December decline. Reddit lost 64 points, Instagram 48, and X 46. In travel, OTAs such as TripAdvisor, Yelp, and Expedia declined, while hotel chains gained. Ray notes that some losers, including Reddit and Indeed, rebounded shortly after the rollout window closed.

Why This Matters

Benchmark these category breakdowns against your vertical. If you work in travel, health, finance, or jobs, Amsive shows which sites gained or lost. This helps distinguish whether the update affected your vertical overall or your site specifically.

The YouTube number is the headline, but it has returned to pre-March levels, making it a correction rather than a new low. The pattern across verticals is more telling. Domains owning the product or service tended to gain, while aggregators or discussion platforms tended to lose.

Amsive interprets Google as favoring “the company that owns the thing” over “the platform to discuss it.” This aligns with the data, but it’s Amsive’s view, not Google’s.

What SEO Professionals Are Saying

Lily Ray, VP of SEO and AI Search at Amsive, wrote on LinkedIn:

“This was a weird core update, but I think the main takeaways are consistent with broader trends we are seeing in Google search: the push toward elevating the actual companies selling the product/service, not the companies ‘writing about them.’”

Read our full coverage: Google’s March Core Update Shifted Visibility Away From Aggregators

Google Preferred Sources Feature Expands To All Languages

Google updated its Search Central documentation to reflect that the Preferred Sources feature is now available in all languages supported by Google Search.

Key facts: Preferred Sources lets users choose publishers they want to see more often in Top Stories and Google Discover. It’s available in all languages supported by Google Search and has added translated downloadable button assets for publishers. The feature works as a user-controlled signal alongside Google’s ranking systems.

Why This Matters

You can use the Preferred Sources button on your site to influence how your site appears in Discover. This expansion is especially important for non-English markets, as the feature was previously English-only, limiting multilingual publishers.

Read our full coverage: Google’s Preferred Sources Is Now A Global SEO Signal

Mueller Says Vibe Coding Won’t Handle Your SEO For You

Google Search Relations team members John Mueller and Martin Splitt discussed vibe-coding websites in a recent episode of Search Off The Record. Both found that AI coding tools could produce functional sites quickly, but getting SEO right still required specific technical direction.

Key facts: Mueller said vague prompts like “add some SEO” lead to vague results. He compared vibe coding to working with a developer who does not specialize in search. The sites he built produced reasonable HTML that would not stand out as vibe-coded. Mueller named Claude Code and Gemini CLI as his current tools.

Why This Matters

Mueller’s experience suggests these tools handle HTML structure and layout well enough, but they don’t make informed choices about canonicals, sitemaps, or crawlability without specific instructions.

Mueller flagged similar gaps in vibe-coded sites before. He reviewed a vibe-coded Bento Grid Generator on Reddit and found issues with crawlability, obsolete meta tags, and content stored in JavaScript files that search engines couldn’t access.

Read our full coverage: Google’s Mueller: Vibe Coding Won’t Handle Your SEO For You

Ask Jeeves Is Gone After Nearly 30 Years Of Search

Ask.com, the search engine that started as Ask Jeeves, shut down. Parent company IAC discontinued its search business as part of a broader refocus.

Key facts: Ask Jeeves was founded in 1996 with a natural-language question format and a cartoon butler mascot. IAC acquired the company in 2005, dropped the Jeeves branding, and by 2010 had shut down the web crawler and outsourced core search. The farewell message on Ask.com closed with: “Jeeves’ spirit endures.”

Why This Matters

Ask Jeeves was the first search engine built on the premise that people should be able to type full questions instead of keywords. That idea did not save the company, but it describes what Google is now building with AI Mode and AI Overviews. The search engine that pioneered conversational search closed the same year conversational search became the industry’s direction.

The closure marks the end of one of the last recognizable consumer search brands from the pre-Google era.

Read our full coverage: Ask Jeeves Is Gone After Nearly 30 Years Of Search

Theme Of The Week: Source Identity Matters More

Every story this week comes back to source identity.

Google adds labels, previews, and signals to links. Amsive’s analysis shows visibility shifting toward brands owning the products or services. Preferred Sources allow users to tell Google which publishers they trust.

If your site is the original source, this week’s signals all point the same way. Google is building more paths back to you. If your site summarizes what others produce, the math is getting harder.

Top Stories Of The Week:

Featured Image: PeopleImages/Shutterstock; Paulo Bobita/Search Engine Journal

Ecommerce MGMT 0 Comments
May 8 2026
WordPress 7.0 Will Ship Without Real-Time Collaboration via @sejournal, @martinibuster

WordPress announced that the real-time collaboration feature will not ship in the upcoming 7.0 release, citing that the current approach is not yet “robust enough” to be included. The response from the WordPress community was a collective sigh of relief.

A post by Matt Mullenweg in the official WordPress Make Slack workspace explained the reasons for removing the feature:

“Hi
I would say I’m not confident with RTC being in core at this point.

I’m fine taking the heat for pulling it out, but the surface area, race conditions, server load, memory efficiency, and the bugs that keep popping up in fuzz tests / etc don’t give me a lot of confidence on our current approach being the robust one we want to support.

There’s also the added burden right now of the WP Engine lawfare and depositions being a time DoS for critical people, perhaps without that in the background we could have gotten to a better place by now, but I don’t think with the current deposition schedule and all the faff around discovery we’re able to support this feature”

Real-time Collaboration (RTC)

WordPress delayed the release of version 7.0 due to the inability to finalize the collaboration feature. Collaboration is Phase 3 of a four phase development road map and version 7 was meant to mark the phase 3 milestone.

Real-time collaboration enables multiple users to edit posts or pages simultaneously. The current default is to lock a web page or post if someone else is working on it. The purpose of collaboration was to bring better editorial workflows by eliminating “post locking” and ensuring changes appear instantly across all active sessions.

Matt Mullenweg tweeted in April that the real-time collaboration feature could also enable AI to edit in real-time, too. That part of the feature was likely to be the most popular use of RTC but aside from Matt’s tweet nothing else was actually said about that.

The commercial WordPress hosting platform WordPress.com has had a controlled roll-out of the RTC feature for VIP and Enterprise users since late 2025.

A support page on WordPress.com explains:

“With real-time collaboration, you and other users on your site can edit the same post or page in the WordPress Editor at the same time and see each other’s changes as you type. This guide covers how to turn real-time collaboration on or off and what to expect while editing together.

This feature is available on sites with the WordPress.com Personal, Premium, Business, and Commerce plans. For free sites, upgrade your plan to access this feature.”

Response To Announcement

Many folks in the WordPress community were not surprised by the delay because it had become evident that finalizing the feature was not progressing fast enough to make the release deadline of May 20, 2026.

Now the process begins to remove the feature from the 7.0 release.

According to the announcement:

“Today, @matt made the decision to remove real-time collaboration from WordPress 7.0 and shared that he is not confident the current approach is robust enough to include in Core at this time, citing concerns around surface area, race conditions, server load, memory efficiency, and recurring bugs found through fuzz testing.

This is a difficult decision, especially given the amount of work that has gone into the feature, but it is being made in service of shipping a stable and reliable WordPress 7.0 release for our users. Work to remove the feature from the release is being organized in #65205 and in the #feature-realtime-collaboration. At this time, the release schedule remains as is and further updates will be provided if the schedule needs to change to unwind this feature.”

Upbeat Response To Removal Of RTC

The general response to the announcement was relief and agreement that it was a good idea. Some disagreed that the feature should ever become a part the WordPress Core because it wasn’t something that most users would need. Although an argument could be made that AI-powered collaboration would be highly popular.

One of the comments to the official announcement on WordPress.org welcomed the decision.

Kento wrote:

“I honestly welcome this decision, besides the fact that this needs further work to confidently get into core, it will also give extension developers more time to prepare, experiment and provide feedback.”

Over on X, WordPress entrepreneur Katie Keith shared her opinion that the feature is delayed but not written off and will likely be returning in the near future.

She shared:

“This looks significant, but I think it’s a delay rather than an actual change of direction or strategy.”

In a completely different discussion on X, Justin Ferriman tweeted:

“That feature always came across as a bit tone deaf. This was the right decision.”

Mark Wesguard agreed:

“Good move. I never understood why this should be in core.”

Hendrik Luehrsen praised WordPress co-founder Matt Mullenweg for his leadership:

“Good call, @photomatt! It’s good to see brave decisions being made in the project at the moment. Inspires hope.”

Similar sentiments were echoed on the Dynamic WordPress Facebook Community (visible to members only), where members expressed the opinion that the feature would be best as a plugin and that the feature was irrelevant for most users.

My own view is that RTC seemed like it had the potential to become a popular way to collaborate in real time with an AI agent. It would also make it easier for teams to collaborate on editing.

Featured Image by Shutterstock/TierneyMJ

Ecommerce MGMT 0 Comments