What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
In a 2026 analysis, Mark Huntley reports that the 100 most-cited third-party editorial pages accounted for about 34.2% of the third-party citation occurrences in the dataset. That is a concentration finding—not proof that those pages caused AI systems to recommend a brand, and not a count of all 41,770 raw citation records.
What the 41,770 figure counts—and what it does not
Huntley’s September 24, 2026, TechBullion article describes an analysis of 135 final consensus research articles across five commercial AI marketing categories. It began with 41,770 raw source-citation records. After processing, the article reports 4,885 unique third-party editorial articles, which generated 29,844 citation occurrences. The 100 most frequently cited individual articles generated 10,199 occurrences—approximately 34.2% of that third-party total. TechBullion’s account of the analysis does not provide row-level records for independent reproduction.
As an Amazon Associate I earn from qualifying purchases.
- 41,770 is the reported raw-record count, not the number of unique pages.
- 4,885 is the reported count of unique third-party editorial articles after processing.
- 29,844 is the reported number of third-party citation occurrences.
- 10,199 is the reported occurrence count for the top 100 individual articles—not the top 100 domains.
The article says its processing excluded fit-review reports to avoid double counting; normalized URLs for protocol, www, trailing slashes, fragments, and common tracking parameters; merged exact-title variants from the same domain; and excluded unresolved Google grounding redirect URLs. These choices matter: counts depend on what was included and how records were grouped.
Why page-level recurrence is a useful clue, not a GEO rule
The concentration suggests that a relatively small group of specific pages appeared repeatedly in the analyzed commercial answers. Huntley also reports that some product reviews appeared in more than 50 consensus articles, with four examples recurring 78, 77, 76, and 73 times. The article excerpt does not identify those reviews or supply the underlying rows, so those examples cannot be checked or attributed to particular publications here.
#1 Best Overall
Repeated appearances can help prioritize which pages to inspect, but a frequently cited URL is not automatically an influential domain, a trusted source in every context, or the reason an answer recommends a product. A citation may support a narrow factual statement; it does not by itself show that the page shaped the answer’s overall judgment. Nor does this one commercial dataset establish a universal pattern across AI systems.
What kinds of external pages appeared
Huntley says the cited third-party material included product reviews, pricing analyses, comparison and alternatives pages, buyer guides, best-of lists, feature breakdowns, and independent evaluations. His interpretation is that company-owned pages may not answer the comparative questions people ask when choosing a product or service. That is a plausible reading of the reported source mix, not a demonstrated causal explanation.
Rank #2
For a company assessing visibility, the practical implication is to inspect the external pages that recur around relevant questions—not just its own content. This “citation neighborhood” can reveal which competitors those pages discuss, what claims they repeat, where third-party descriptions conflict with company information, and which recurring sources omit the company.
Audit AI visibility with separate signals
Do not compress every result into a single “visibility” score. A brand can be mentioned without being recommended; a page can be cited without the brand being endorsed; and a brand can be absent from a prompt where it would be a plausible option. Track those outcomes separately, and distinguish owned evidence from third-party evidence.
Rank #3
- Build prompt groups around real intent. Separate broad informational questions from high-intent buying questions. For example, “What is a medical alert system?” asks for an explanation; “What is the best medical alert system for a senior living alone?” asks for a recommendation under a particular use case. Huntley presents such prompts as examples, not as a measured frequency study.
- Record the outcome for each prompt. Mark whether the company is mentioned, recommended, cited as a source, or absent. Keep the prompt wording and date with each observation because outputs can change.
- Capture the cited evidence. Record individual URLs as well as domains. A domain-level summary can hide that one specific article—not the publication generally—accounts for repeated citations.
- Inspect recurring pages and competitors. Note which products and companies are included, what claims are made, and whether pricing, features, limitations, or other facts conflict with current company information.
- Compare owned and external gaps. Identify unanswered questions on the company’s own site, then separately identify recurring third-party pages that omit or misrepresent the company. Treat these as different problems: publishing a page does not guarantee that an external source will change its coverage.
- Recheck over time. Keep informational and commercial prompt groups separate and compare observations across dates. Treat a citation count as a monitoring signal, not a promise of future recommendations.
What Google’s guidance means for GEO
For Google Search specifically, Google Search Central says AI Overviews and AI Mode remain grounded in core Search ranking and quality systems. Its guide describes generative features using retrieval-augmented generation to draw relevant current pages from the Search index and query fan-out to generate related searches. Google’s guidance is not a description of how every answer engine works.
Google recommends foundational SEO, useful and distinctive content, and meeting Search eligibility and crawlability requirements. Its guide says site owners do not need an llms.txt file or other special AI files, artificial content chunking, rewrites made solely for AI, inauthentic mentions, or special structured data to gain visibility in Google’s generative search features. Structured data may still support rich-result eligibility, but Google says it is not required for generative AI search. The guide also points site owners to the Generative AI performance report in Search Console and cautions that third-party tools do not have access to Google’s internal ranking or AI systems. See Google Search Central’s guide to AI features and your website, last updated July 10, 2026. Google announced the guide on May 15, 2026, describing its coverage of useful content, local, shopping, image and video content, misconceptions about AEO and GEO, early AI-agent guidance, and SEO’s continuing relevance: Google’s announcement.
Google’s own summary is: “From Google Search’s perspective, optimizing for generative AI search is optimizing for the search experience, and thus still SEO.” The useful distinction is platform scope: this is Google’s guidance for Google Search, not evidence that every AI service uses the same retrieval or ranking process.
How much confidence to place in the headline finding
The TechBullion article describes its dataset and processing, but does not publish row-level data, a full measurement instrument, platform-by-platform counts, confidence intervals, or an independent audit. It therefore supports a carefully attributed report of what this dataset found; it does not establish a causal link between citations and recommendations or prove that the same concentration holds across the market.
A separate paper by Mahe Chen, Xiaoxuan Wang, Kaiwen Chen, and Nick Koudas, dated September 10, 2025, reports differences between AI search systems and traditional Google search in earned-media sourcing, domain diversity, freshness, cross-language stability, and sensitivity to query phrasing. It is separate context, not validation of Huntley’s 41,770-record analysis: arXiv paper abstract.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
