Google Expands Reporting Beyond Websites

The Visibility Report #28 | August 4, 2026

Google has rolled out platform properties globally, giving publishers a clearer way to connect social and video channels with search performance. That matters because visibility is no longer confined to pages on your domain: now you have to win the ranking, the citation, and increasingly the platform mention.

What matters: AI search visibility is becoming cross-surface visibility, not just website visibility.

Operator read: measure websites, citations, social/video properties, product-feed data, ad surfaces, and generated answer mentions separately.

Watch the trade-off: controls that limit AI search use can also affect ordinary search presentation or discovery.

Google Rolls Out Platform Properties Globally

Google has expanded platform properties globally and released a new guide to measuring social and video performance. The change gives search teams a more direct way to account for content published outside their primary websites. It also reinforces a broader shift: Google increasingly treats websites, social profiles, and video channels as parts of the same discovery system.

Operators should review which brand-owned profiles can be associated with their properties and decide who owns reporting across SEO, social, and video. The announcement does not resolve attribution across AI answers, organic results, and platform engagement, so teams will still need their own reporting model. (Because one more dashboard was apparently inevitable.)

📎 Google Search Central​

Google's AI Search Opt-Out Choices Have Trade-Offs

Search Engine Journal examined what publishers can currently do if they do not want content used in Google's AI search features. The central problem is that controls affecting AI-generated results may also affect ordinary search presentation or discovery, depending on the mechanism used. Publishers therefore face a business decision rather than a clean AI-only switch.

Teams should document which controls they use, what traffic those controls may affect, and whether restricting AI use is worth the possible loss of search exposure. The rules and interfaces may continue to change, so this probably belongs in a recurring technical review rather than a one-time policy decision.

📎 Search Engine Journal​

Zero-Click Search Looks Different by Destination

iPullRank analyzed 13 billion Google searches and found that nearly half ended without a click, according to its report. More importantly, the effect was not distributed evenly: Google absorbed about a third of Wikipedia's potential traffic while having a much smaller effect on Reddit. The analysis also reports that ChatGPT now receives a larger share of its Google clicks from ads than other leading destinations in the dataset.

The operator takeaway is to stop treating zero-click rate as a universal benchmark. Compare performance by query type, destination, brand status, and result feature, then measure whether visibility produces later branded searches or conversions. Analytics aren't truth. They're opinions with decimal points.

📎 iPullRank​

OpenAI Appears to Be Testing Agent-Based Ads

Search Engine Land reports signs that OpenAI is developing chatbot-native ads capable of launching AI agents. If released in that form, an ad could move beyond sending a visitor to a landing page and instead begin a task inside the conversation. That would change what advertisers measure, because an agent action may matter more than a conventional click.

The report concerns a product still under development, so formats, controls, and launch timing remain uncertain. Search and paid media teams should nevertheless start defining how they would evaluate an AI-assisted action, including attribution, consent, lead quality, and what happens when the agent gets something wrong.

📎 Search Engine Land​

AI Shopping Visibility Starts With Product Feeds

A Search Engine Land analysis argues that product feeds, rather than product pages alone, determine whether AI shopping systems can interpret and recommend an item. Titles, identifiers, availability, pricing, categories, and other structured attributes give these systems the data required to compare products. A polished page cannot compensate for an incomplete or inconsistent feed when the shopping engine relies on structured inputs.

So commerce teams should audit the feed as a source of truth, not just an advertising export. Compare feed claims with landing pages, test how products appear in conversational shopping results, and prioritize corrections that affect eligibility or factual accuracy.

📎 Search Engine Land​

Profound Maps the Sources Behind AI Citations

Profound examined where AI citations originate, giving visibility teams a way to look beyond their own domains when planning content distribution. Citation sources can reveal which publishers, communities, reference sites, and content formats an answer engine trusts for a particular topic. That makes source analysis useful for digital PR and partnerships as well as on-site optimization.

Vendor research should be treated as directional until the sample, prompts, markets, and collection methods are clear. Use the findings to build a test list, then check the actual sources appearing across your highest-value prompts and engines.

📎 Profound​

Digital PR Is Becoming an AI Visibility Input

Semrush outlines five digital PR tactics intended to increase the likelihood that brands appear in AI-generated answers. The premise is practical: answer engines often rely on third-party sources, so coverage and corroboration can matter alongside content published on a brand's website. Measurement needs to connect placements with later citations, mentions, sentiment, and prompt-level visibility rather than counting links alone.

The article is vendor guidance, not proof that every placement causes an AI citation. Teams should test campaigns by recording a baseline prompt set, tracking new coverage, and checking whether citations change after answer engines have had time to discover the sources.

📎 Semrush​

Rankings Alone Do Not Measure AI Visibility

Lumar explains how GEO adds answer inclusion, citations, and entity understanding to the traditional ranking model. Rankings still influence discovery, but they do not show whether an AI system used a page, cited a competitor, or mentioned a brand without linking. That creates a measurement gap for teams relying only on rank tracking and organic sessions.

A useful next step is to pair existing SEO metrics with a stable prompt set, citation capture, crawler access checks, and brand-mention monitoring. The framework is an overview rather than independent evidence about individual ranking factors, so test each recommendation before treating it as a rule.

📎 Lumar​

More From This Week

Search & AI Visibility

• ​Why Every SEO Team Now Needs a Social Topical Map (Search Engine Land) - Connects social publishing with entity coverage, demand discovery, and search visibility.
• ​AI Makes SEO Faster, but Human Expertise Still Wins (Search Engine Land) - Examines where automation helps SEO production and where judgment remains necessary.
• ​Social Search Data for All and AI-Detected Pages Rank Lower (Search Engine Journal) - A roundup of social search data and reported correlations involving AI-detected pages.

From the Tool Blogs

• ​Brand Affinity Is Live in SparkToro Reports (SparkToro) - Audience reports now identify brands, companies, and products that can inform entity research and AI citation hypotheses.
• ​What Is Schema Markup? (Semrush) - An implementation guide to structured data that helps search and AI systems interpret page entities and attributes.
• ​Three Workflows to Try With the Profound MCP (Profound) - Examples for connecting AI visibility data with repeatable research and reporting workflows.
• ​Profound Introduces Brainstorm Mode in Aim (Profound) - A new mode for generating and evaluating ideas using visibility data.
• ​The Shortlist Is the New Shelf (Profound) - Argues that inclusion in AI-generated consideration sets is becoming a key brand visibility metric.
• ​GEO vs. SEO: What's the Difference? (Peec AI) - An overview of where AI answer optimization extends traditional search work.

Agency & Practitioner Insights

• ​Canva Built a 90-Million-Backlink Empire. Does It Survive AI Search? (Foundation Inc) - Examines whether Canva's search-led content model remains defensible when answers replace some clicks.
• ​The AI SEO Guide: From Concepts to Application (Search Influence) - A practical introduction to adding AI visibility work to an existing SEO program.

Engine Announcements

• ​Five Ways AI Mode Helps With Real-World Tasks (Google) - Product examples show the query types and task-oriented behavior Google wants users to adopt in AI Mode.

So What Do You Do About It?

Pick ten commercially important questions and record what appears across Google organic results, AI Overviews, AI Mode, Gemini, ChatGPT, and Perplexity. Capture the cited sources, mentioned brands, product-feed details, ad surfaces, and social or video properties separately. Repeat the test each week for a month; the differences between surfaces will probably be more useful than any single visibility score.

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