AI Search Citation Study
Description
This dataset records the public AI search visibility results behind Paloren's citation study. It contains 60 branded and commercial AI-consulting queries checked in Google search on 27 September 2026, with 22 citation records from the result set. The files separate domain and URL observations, boolean AI Overview mentions, and SERP feature counts for 30 core queries. Analysts can use the files to review source diversity, citation patterns, and feature mix without collecting search results again. The package includes a README describing the method, column definitions, licence and suggested citation. Attribution must credit Paloren under CC BY 4.0.
Files
Steps to reproduce
We defined 60 branded and commercial AI-consulting queries, then collected Google search results on 27 September 2026. For each result we logged its rank, domain and URL. AI Overview citation mentions were encoded as booleans. For the 30-query core subset, AI Overviews, discussions, local packs and related searches were counted as SERP features. Observations were normalized into three CSV files: citations.csv, aio_mentions.csv and serp_features.csv.