A high volume of citations in artificial intelligence search engines does not guarantee a brand will win customer recommendations, according to a major study examining corporate visibility across generative AI platforms.Research published in Leoprd's Reputation to Revenue 2026 report reveals that brands chosen as primary recommendations by AI engines actually averaged fewer citations than their passed-over competitors—8.4 citations compared to 9.2, respectively.The Recommendation GapThe findings highlight a sharp divide between AI visibility and commercial impact. Across unbranded recommendation prompts, monitored brands were selected as an AI platform's first choice just 29% of the time. They were mentioned but passed over in 35% of instances, and omitted entirely in 36%.The study analyzed 31,200 AI answers from a broader dataset of 214,000 responses spanning 27 brands, seven categories, and eight AI platforms across Australia and Great Britain.The performance gap was particularly pronounced in specific sectors:• Travel & Transport: Brands appeared in 90% of relevant answers, but earned the top recommendation only 28.1% of the time.• Fintech: Brands achieved a 92.3% presence rate, but were selected as the primary choice in just 32.1% of queries.Quality of Data Trumps Citation VolumeWhile communications discussions around generative engine optimization (GEO) have heavily focused on making brands searchable and citable, the study indicates that citation source mixes—comprising commercial, editorial, and review outlets—were virtually identical for winning and losing brands.Instead, recommendation outcomes hinged on the specific messaging contained within those sources and whether the data gave the model a compelling rationale to prefer one company over another."If you think about traditional metrics, that's telling you how loud you are, whereas the new ones will tell you whether you're chosen or not," said Amber Morris, APAC Head of Communications at Pinterest, in the report. "They're going to tell you the relevance and impact of what you've done."AI's Role in Pre-Purchase DecisionsThe shift in how search algorithms evaluate brands comes as consumer reliance on AI for decision-making grows. In a separate analysis of more than 420,000 anonymized real-world prompts across the U.S., U.K., and Australia, Leoprd found that 44% of queries asked AI platforms for research or advice, while 23% directly requested comparative evaluation or product recommendations.Furthermore, consumers utilized AI tools approximately 20 times more frequently before making a purchase than after, with pre-purchase research representing 27% of all analyzed prompts compared to just 1.3% for post-purchase support.The data suggests that shaping how AI engines evaluate brand authority before a consumer visits a company website has become a critical reputational and commercial priority, superseding basic search engine visibility.(Photo by Igor Omilaev on Unsplash)