Original research on AI visibility
We run the same measurement Beacon sells — real buyer questions, sampled across AI engines, checking who actually gets named — and publish what we find, including when the numbers aren’t flattering.
When clients ask AI for a Swiss private bank, who gets named?
48 buyer questions, 3 engines, 23 public banks. Gemini names a bank in 42 of 48 answers; Perplexity in 32. One question cluster goes to almost no bank at all.
When clients ask AI for an independent wealth manager, who gets named?
100 buyer questions, 26 boutique & independent firms. 67 of the 100 questions got no named firm on any engine — the category is almost entirely unclaimed in AI answers.
When U.S. clients ask AI for a wealth manager, who gets named?
99 buyer questions, 26 wirehouses, private banks and independent RIAs. The U.S. companion to our Swiss studies — same method, same rigor, a much bigger market.
When buyers ask AI which business software to buy, who gets named?
100 buyer questions, 24 CRM, HR, ERP and support vendors. Only 5 of the 100 questions got no named vendor on any engine — the most crowded category we’ve measured, with Salesforce and HubSpot dominating.
When LPs ask AI for venture capital fund due diligence, who gets named?
100 LP due-diligence questions, 23 venture capital funds. 64 of the 100 got no named fund on any engine — due diligence and fee questions are almost entirely unclaimed, the widest-open category we’ve measured.
When buyers ask AI which DevOps vendor to pick, who gets named?
100 buyer questions, 23 cloud, CI/CD and observability vendors. Every single question named a vendor on at least one engine — the most crowded category we’ve measured. AWS, Azure and Google Cloud dominate; Gemini names a vendor 100 of 100 times but grounds on nothing citable.
When buyers ask AI which cybersecurity vendor to pick, who gets named?
100 buyer questions, 24 EDR, identity, cloud security and compliance vendors. Only 1 of 100 questions got no named vendor — Microsoft Defender and CrowdStrike dominate, with Perplexity leaning almost entirely on independent research sites rather than vendors’ own pages.
When families ask AI for a family office, who gets named?
100 buyer questions, 26 multi-family offices and bank-run practices. Global, not one market — the US, UK/Europe and Switzerland, since family offices and their clients are inherently cross-border.
When Zurich & New York patients ask AI for a dentist, who gets named?
Two location-dedicated studies — same 100 patient questions, the city swapped, so Zurich and New York are directly comparable. Solo practices, group clinics and implant / ortho specialists, all in one study per city.
When buyers ask AI which insurer to use, who gets named?
Two region-dedicated studies — same six clusters and the same buyer intent question by question, phrased for each market, so Europe and the U.S. are directly comparable. Direct carriers, comparison platforms and digital-first insurtechs, all in one study per region.