Lantern Releases New Research Showing AI Shopping Recommendations Are Far More Stable Than Daily Tracking Suggests

Study of more than 3,300 AI shopping responses finds that most day-to-day movement reflects model variability rather than meaningful changes in brand visibility

NEW YORK, July 28, 2026 /PRNewswire/ -- Lantern today released new research examining how consistent AI shopping recommendations actually are over time, finding that the brands AI recommends remain remarkably stable despite the daily fluctuations many marketers observe in AI visibility dashboards.

As AI becomes an increasingly important channel for product discovery, brands have begun tracking how often they appear in AI-generated shopping recommendations. Daily changes in visibility scores often lead marketers to believe that AI recommendations are constantly shifting.

Lantern's research suggests the opposite. While large language models rarely produce identical responses twice, the brands they recommend and the positions they occupy remain largely consistent over time. Much of the day-to-day movement reflects the inherent randomness of generative AI rather than meaningful competitive change.

The research tracked how approximately 3,300 AI shopping recommendation trends changed over time across 186 combinations of brands, consumer prompts, and AI models. Rather than analyzing a single set of AI responses, Lantern repeatedly queried the same real consumer shopping prompts over an extended period, including queries such as "best bath towels" and "best cordless vacuum for pet hair," measuring whether brands appeared, where they ranked, and how favorably they were described. Of the combinations analyzed, 132 involved unbranded category searches where brands compete directly for AI recommendations.

"One of the biggest misconceptions about AI commerce is that recommendations are constantly changing," said Andrew Lissimore, CEO and co-founder of Lantern. "Large language models naturally generate slightly different answers every time you ask the same question, and that creates the impression that rankings are moving more than they actually are. Our research shows the underlying competitive landscape is much more stable than the day-to-day noise suggests."

Among the study's findings:

• Brand visibility remained highly consistent. Whether a brand appeared in an AI answer remained unchanged across the tracking period in 80% of all brand-prompt-model combinations. Among competitive unbranded prompts, 71% remained stable.
• Top recommendations rarely changed. Brands holding the number one recommendation maintained that position in roughly two-thirds of tracked combinations.
• Daily fluctuations did not translate into long-term movement. Visibility scores oscillated around stable baselines with no evidence of sustained upward or downward trends. In practice, a score from several days earlier predicted current performance about as well as the previous day's score.
• Many competitive opportunities remain available. Across unbranded shopping prompts, tracked brands appeared in roughly two-thirds of AI responses, leaving a meaningful share of recommendations without established brand representation.

Lissimore added, "That distinction matters because brands shouldn't build AI commerce strategies around daily fluctuations. Instead, they should focus on the long-term factors that consistently influence whether their products are recommended, rather than reacting to every small movement in a dashboard."

About Lantern

Lantern is an AI commerce platform that helps brands drive product visibility inside AI-driven shopping. The platform predicts how products are surfaced across AI systems and applies changes that improve performance. For more information, visit https://lantern.is/

View original content:https://www.prnewswire.com/news-releases/lantern-releases-new-research-showing-ai-shopping-recommendations-are-far-more-stable-than-daily-tracking-suggests-302836506.html

SOURCE Lantern