A small independent test suggests that searching for a major brand’s name can now bring up an AI-generated explanation alongside the familiar company website. Chris Long reported AI Overviews for 93% of the brands he tested. The finding raises a practical question for brand teams: what does Google tell people about the business before they reach its own pages?
What the 93% figure actually measures
Search Engine Roundtable covered the observation on September 30, 2026, describing a sample of roughly 100 brands. Barry Schwartz reported checking several examples himself and seeing similar results. This provides partial corroboration of the behavior, rather than an independent reproduction of the entire percentage.
The report shows variation in placement. Many examples put the AI Overview below the company’s main result or further down the page; an Adobe example places it at the top. The 93% figure applies to the tested brand set. It is not a population estimate for all branded searches, a Google rollout announcement or evidence that 93% of users see these modules.
A brand query can serve more than navigation
Someone typing a company name may want its website, but may also be checking what it sells, whether it serves a particular market or how its products relate to one another. An AI summary can become another place where those questions are answered. That makes the wording of the summary relevant even when the official website remains easy to find.
Placement matters when assessing the possible effect. An answer above the main company result creates a different reading sequence from one further down the page. Neither configuration, on its own, establishes a loss of clicks, a change in trust or an effect on revenue. Those outcomes require separate evidence.
What GEO means for a company’s own name
For this story, generative engine optimization is best understood as examining how a brand is represented in generated answers. It includes checking factual accuracy, the sources presented and whether the explanation reflects the company’s current products and identity. It cannot give a company control over Google’s wording.
A useful audit would record the exact query, date, language, location and device, then capture the answer and its position relative to the official result. Review the claims and linked sources. Distinguish a factual error, such as an outdated product description, from a framing choice that the marketing team simply dislikes. That distinction helps prioritize useful corrections.
Start with reliable information and ordinary SEO
Google’s guidance for AI features says established SEO practices remain relevant and that there are no special additional optimizations required for AI Overviews or AI Mode. A page must be indexed and eligible for a search snippet to qualify as a supporting link. Eligibility does not guarantee selection.
Our practical recommendation is to review the pages that explain the company: its main description, product documentation and material announcements. Make important facts clear, current and consistent. If a generated answer exposes an outdated statement on an owned page, fix that source and monitor subsequent results; do not assume the next answer will immediately change.
Measure representation alongside traffic
Keep a recurring record of answer presence, placement, factual accuracy and linked sources. Compare it with branded search performance, while avoiding a causal claim from simultaneous changes alone. This limited observation supports closer monitoring of brand answers. It does not establish that every company faces the same exposure or that a particular optimization will secure a citation.