How do you measure which brands AI assistants name? We carry out the experiments, with pre-specified metrics, confidence intervals and thousands of archived answers, and publish what we find, including the parts that are inconvenient. GenSig's own methodology is built on these results.
We asked the same purchase-intent questions 2,500 times across five API configurations and 30 times by hand in the consumer apps (ChatGPT web, Gemini web). The bare LLM API missed 3 of the 5 brands the app sessions consistently name; temperature 0 made answers highly repeatable, but with the wrong brand set; and each search-grounded API turned out to mirror its own consumer surface.
Every study ships with its raw data and code: download the Research #1 dataset (2,500 API responses + 30 consumer-app sessions, prompts, brand dictionary and analysis code).
New to AI visibility? Start here: brand visibility in AI: what it is, why it matters and how to measure it, then the practical measurement guide.
More studies are on the way, replicating this one across categories and markets. Questions, corrections and replication requests: [email protected].