Visibility #43: Google Search Is Querying More Than Your Query


The Visibility Report​
Issue #43 | Week of November 9 - November 15, 2026

Google Search Is Querying More Than Your Query

Google's newest Search model can split one complicated question into several related searches, then build an answer from what it finds. That makes the pages behind the answer harder to predict, but it does not make the work mysterious.

​Subscribe to The Visibility Report for a source-checked briefing every Tuesday.

Google expands query fan-out with Gemini 4 Hydrogen

Google says Gemini 4 Hydrogen is now available globally in Search. The model upgrade is designed to handle longer, multi-part questions, follow instructions more closely, and produce more visual responses.

The part marketers should study is query fan-out. Google says Hydrogen runs simultaneous searches across the subtopics it identifies in a question, then connects the user to a broader range of links. A buyer may ask one question, while Google searches several concepts that were never typed.

The practical response is not to write a page for every possible prompt. Map the real follow-up questions around a decision. Make product details, comparisons, limitations, evidence, and next steps clear enough to stand on their own. Then check which parts of that decision Google sources from you, competitors, communities, and independent publishers.

​Source: Google's November 12 announcement of Gemini 4 Hydrogen in Search​

United separates search indexing from AI agents

United Airlines updated its terms to let conventional search engines index public pages while requiring written permission for automated agents that search, book, buy, manage, or collect information on its site or app. The terms also restrict using United's fares, schedules, availability, pricing, and other site data to train, fine-tune, ground, or evaluate an AI model.

That distinction matters. Search indexing, answer grounding, and an agent taking action are different uses of the same website. A robots.txt file alone does not express all three decisions. Legal terms, technical controls, authenticated access, and approved data feeds may each have a role.

This is one company's policy, not a standard every site should copy. It is a useful prompt for a cross-functional review: which automated uses help customers, which expose sensitive actions or data, and where does permission need to be explicit?

​Source: Search Engine Journal's review of United's November 11 terms and announcement​
​Source: United Digital Trust Center and customer privacy resources​

Reddit turns product conversations into review pages

Reddit is testing Reddit Reviews with a subset of US users. The product organizes comments from across communities into review pages with opinion summaries, recurring themes, related products, source comments, communities, and, where available, product information and purchase links.

The initial test covers selected products in electronics, health and beauty, home and garden, and apparel. Reddit says reviews link back to the original discussions. It also says brand-owned accounts are excluded from influencing review summaries and repeated contributions from the same account are limited.

For brands, start with the source material. Review the comments behind a summary. Look for repeated product problems, misunderstood features, and comparisons buyers keep making. Fix the product or the explanation before trying to influence the summary.

​Source: Reddit's product announcement and follow-up on safeguards​

Publishers test a collective license for AI answers

The Internet Licensing Collective launched with more than 3,000 media brands, content owners, and publishing platforms. Yahoo Scout is its first announced commercial licensee and will pay participating publishers when their content is used in AI answers.

This does not settle the argument over licensing, attribution, or payment. It does create another route for publishers that do not have the scale to negotiate one agreement at a time. The questions to watch are how usage is measured, how money is allocated, and whether licensed material receives visible attribution.

​Source: Search Engine Roundtable's report on the ILC launch and Yahoo Scout agreement​

From Practitioners

  • ​iPullRank: An analysis of Search Console ranking data across 166 brands. The useful reminder is methodological: control for authority and other variables before treating a simple correlation as a cause.
  • ​Foundation: A study of YouTube citations in one project-management prompt set. It is a category-specific sample, not a universal benchmark, but it shows why video transcripts, titles, and clear demonstrations deserve a place in citation audits.
  • ​A GrowthHacking community thread: The soundest advice in a noisy discussion was simple: test real buyer questions, record the brands and sources that appear, and compare those pages with your own. Start with observation before buying a score.

From the Tool Blogs

More From This Week

What to watch

  • Which subqueries Hydrogen runs for complex buying and local-service questions.
  • Whether more companies publish separate rules for indexing, grounding, and agents that take action.
  • How Reddit keeps commercial manipulation out of review summaries as brands learn how the system works.
  • Whether collective licensing produces transparent usage and payment reports.

What to do next

  • Choose five important buyer questions and record the sources Google cites for each subtopic, not just the final answer.
  • Write down your policy for search crawlers, AI retrieval, training, and agents that can change data or spend money.
  • Read the source comments behind any Reddit Review that covers your product or category.

One thing to do this week: audit one complicated customer question as a set of subquestions, then see whether your site earns a useful place in each part of the answer.

The Visibility Report | Will Scott​
This newsletter is produced collaboratively by Will Scott and Bob, an AI agent. Human oversight, AI efficiency.
​Subscribe at thevisibilityreport.com​

The Visibility Report

Join 500+ digital marketers getting weekly AI visibility tools, tactics, and search updates. Free • 5-min read • No spam.

Read more from The Visibility Report
A glossy white robot places a cyan verification marker into a brass sidewalk locator outside a warmly lit storefront while three arrow cards point toward the business.

The Visibility ReportIssue #42 | Week of November 2 - November 8, 2026 ChatGPT Webmaster Tools May Be Coming OpenAI appears to be testing something search marketers have wanted for years: a first-party view of how businesses appear in ChatGPT. Think of it as ChatGPT Webmaster Tools, with one important caveat. This is an unconfirmed, unfinished feature seen by a third party, not an OpenAI product announcement. Subscribe to The Visibility Report for a source-checked briefing every Tuesday....

A glossy white robot in a warm print studio tears a permission sheet into separate discovery and AI-training choices.

The Visibility ReportIssue #41 | Week of October 26 - November 1, 2026 Publishers Get a Clearer Choice About AI Training Publishers have been stuck with a blunt choice: let AI companies use their content or risk disappearing from AI answers. New controls from Google, OpenAI, and Microsoft make that choice a little less blunt. Microsoft draws the clearest line between model training and grounding, which is the use of current web content in generated answers. A publisher can say no to training...

A human-scale white robot stands beside a much taller lighthouse as its beam guides boats toward a harbor.

AI Traffic Stays Small as AI Answers Reach Buyers The Visibility Report #40 | Week of October 20-26, 2026 AI assistants still send very little identifiable referral traffic compared with Google. Marie Haynes found that ChatGPT sessions equaled 0.62% of Google organic volume across 100 sites and more than 145 million sessions. The same week, OpenAI explained how ChatGPT ads use a separate language model to read multi-turn intent, retrieve candidates, and run an auction. Those findings put the...