SEO »
Mentions matter more than citations, and narrow leads often change, revealing the instability of AI search brand authority, Semrush finds. #
Only 15.2% of ChatGPT topic categories had a clear brand owner across related buyer questions, according to a Semrush analysis.
That means about 85% of categories had no dominant brand. Appearing in one prompt didn’t mean a brand appeared consistently across related questions in the same topic.
How ownership was defined. Semrush defined a category owner as the brand with the highest share of mentions, appearing in at least four of five related prompts and leading the runner-up by at least 5 percentage points.
- Only 15.2% of the 1,094 categories met that standard.
- Another 31.2% had an emerging leader, with one brand appearing in at least three prompts but without the required lead.
- The remaining 53.7% were unsettled, with no brand appearing in at least three of five prompts.
Popular topics,** few clear owners. **Semrush split categories into two groups by AI search demand. Only 11.3% of topics in the top half had a clear owner, compared with 19% in the bottom half.
- The top half accounted for 98% of the AI search volume in the sample.
- Most ChatGPT demand Semrush tracked was in categories without a clear brand owner.
Broad SEO metrics had limited predictive value. Traditional SEO strength didn’t reliably predict which brands owned ChatGPT topics. “Traditional SEO metrics aren’t enough to explain who owns a topic. While they play their role, there’s more to it,” wrote Kevin Indig, founder of Growth Memo.
- Semrush compared category owners with runners-up across branded search volume, organic traffic, and Authority Score.
- Owners had higher branded search volume in 55.7% of comparisons, higher organic traffic in 48.4%, and a higher Authority Score in 52.5%. Branded search volume was the only metric that reached statistical significance.
Most-cited rarely meant most-mentioned. Semrush measured ownership by brand mentions in ChatGPT’s answer text, not by the links cited below.
- Mentions and citations often didn’t align. Only 21% of the most-cited domains in a category were also the most-mentioned brand.
Why we care. Semrush found that prompt-level visibility didn’t always carry across related buyer questions. That suggests topic-level visibility is more useful for assessing whether ChatGPT consistently associates a brand with a category.
Brands that owned a topic usually kept it. Clear category owners stayed in first place in 90.4% of month-over-month comparisons. Weaker leaders changed more often, with the top brand switching in 1,950 of 5,470 comparisons across emerging and unsettled categories.
**Narrow leads were less stable. **When a top brand later lost first place, its typical lead was just 1.3 percentage points. When it stayed on top, its typical lead was 2.9 points, more than twice as large.
- Semrush said the data didn’t explain why leaders changed. It showed only that brands with smaller leads were more likely to lose first place.
About the data. Semrush and Kevin Indig analyzed 1,094 U.S. categories in ChatGPT from January through June 2026 using the Semrush AI Visibility Toolkit. The dataset covered more than 50,000 brands, 220,000 domains, 600,000 citations, and 220,000 URLs. Each category included five prompts tied to common buyer questions, including definitions, comparisons, alternatives, use cases, and buying decisions.
The study. AI visibility is a topic-level game: A study of 50,000 brands in ChatGPT [Study] Search Engine Land is owned by Semrush. We remain committed to providing high-quality coverage of marketing topics. Unless otherwise noted, this page’s content was written by either an employee or a paid contractor of Semrush Inc.