Victorious Study Finds AI Recognizes 96% of Brands but Omits 89%
Victorious' Q2 2026 report says eight AI systems accurately described 96% of 175 brands when prompted by name, while 89% of the same cohort never appeared in category and buyer-oriented answers. The vendor-produced study covered five industries and found recognition and selection behaved as distinct metrics, but its correlations do not establish that off-site mentions cause inclusion in AI responses.
Victorious' Q2 2026 Quarterly Search Report says AI systems accurately described 96% of 175 tested brands when asked about them directly, while 89% of the same brands never appeared in answers to broader category and buyer-oriented questions.
The SEO agency tested brands across legal services, healthcare, SaaS, financial services and ecommerce/retail. For the recognition test, it asked ChatGPT, Claude, Gemini, Copilot, Perplexity, Google AI Overviews, Google AI Mode and Meta AI to describe each brand, then graded the answers against the brand's own website as correct, vague, outdated, wrong or not recognized.
Recognition and selection measure different behavior
Victorious separately measured whether the brands appeared in standardized category prompts. A named-brand question supplies the entity to the model; a category question requires the system to select a limited set of companies. The 96% recognition rate and 89% nonappearance rate therefore describe different tasks rather than conflicting results.
The report also examined which web signals moved with AI mention rates. It found the strongest measured correlations for referring domains (r = 0.49) and third-party web mentions (r = 0.45). Those are associations within this dataset, not proof that acquiring a link or mention causes an AI system to name a brand.
How to use the result
Teams measuring AI visibility should keep at least two metrics separate: whether a system can accurately describe a brand when named, and whether it selects the brand for realistic category, comparison or shortlist prompts. A single blended visibility score can hide the gap between factual recognition and commercial discovery.
The research was produced by Victorious, which sells SEO services, and the retrieved material does not describe independent replication or external auditing. Platform behavior, retrieval systems and prompt wording can also change over time. The findings are best treated as a vendor-produced benchmark and methodology prompt, not a universal threshold or causal prescription for marketing spend.
Key Points
- 1Victorious says eight AI platforms accurately described 96% of 175 brands when prompted directly, while 89% of the cohort never appeared in category answers.
- 2The study treats direct brand recognition and selection for broader buyer questions as separate behaviors that should be measured independently.
- 3Referring domains and third-party mentions correlated with mention rates in the dataset, but the vendor-produced analysis does not establish causation.
Scoring Rationale
The report gives search and marketing teams a useful way to separate brand recognition from selection in AI-generated category answers and publishes concrete cohort and correlation figures. Its evidentiary weight is limited because the research is vendor-produced, was not independently replicated in the retrieved material, and reports correlations rather than causal effects.
Sources
Primary source and supporting public references used for this report.
Practice interview problems based on real data
1,625 SQL & Python problems across 15 industry datasets — the exact type of data you work with.
Try 250 free problems
