AI Search Favors Some Brands Over Others, Study Finds

An analysis conducted by Fractl has revealed that artificial intelligence search models exhibit a notable bias in brand recommendations, with some brands enjoying significant visibility despite weaker traditional search engine optimization (SEO) footprints, while others with strong SEO presence are largely absent from AI-generated answers. The study examined how consistently AI models recommend brands, identifying those that disappear despite robust SEO, and pinpointing signals that differentiate overperforming brands from those that under-index. While over 90% of brands in the dataset behaved as expected, with stronger traditional search authority correlating with greater AI visibility, the outliers present a more complex challenge for digital visibility. These AI answers often feature default brands, meaning some companies with high domain ratings, extensive keyword portfolios, and millions of monthly organic visits were barely surfaced within their respective categories. Conversely, brands with smaller traditional search footprints appeared far more frequently than their SEO metrics would suggest. Across all industries reviewed, a consistent pattern emerged: a select few brands surfaced repeatedly, irrespective of prompt phrasing. These default recommendations were not exclusively the largest companies, nor necessarily those with the most traffic, nor those topping traditional SEO leaderboards. The travel industry demonstrated the highest concentration, with Booking.com (285 mentions), Airbnb (227 mentions), and Expedia (215 mentions) collectively accounting for approximately 20% of the sector’s total brand mention volume in AI responses. This indicates that three brands alone were responsible for one in every five recommendations within travel. In HealthTech, Teladoc emerged as a dominant leader with 275 mentions, significantly outpacing its closest competitor, Amwell, which garnered 220 mentions, a difference of roughly 25%. This suggests a settled preference within AI models for certain providers in this sector. The wellness category, however, presented a more diverse landscape, featuring a deeper bench of frequently recommended brands. Peloton, Headspace, Calm, Whoop, and Oura each surpassed 168 mentions, indicating no single brand holds a monopolistic position. This contrasts with the more concentrated industries, suggesting a broader range of acceptable recommendations within wellness. The findings underscore a shift in how brands achieve visibility, moving beyond traditional SEO metrics to factors influencing AI model selection, which could necessitate new strategies for brands aiming to maintain or increase their presence in AI-driven search environments.