The State of AI Search Citations

What the Report Answers

Six questions every marketing leader will be asked this year. The report answers them with data, not opinion.

We’re running a single coordinated research program across ten B2B verticals. The output is not a vendor-funded survey. It is a primary-research benchmark with batch IDs, raw data, and reproducible methodology that any reader can verify.

• Citation Concentration Which 20 domains receive 40% of the citations in each vertical? Where does your category’s authority actually live?
• Engine Drift How much does the citation set shift between Claude, ChatGPT, Perplexity, Google AI Overviews, and Microsoft Copilot on the same buyer question?
• Persona Sensitivity How much does ICP framing change what gets cited? Per vertical, per engine, with confidence intervals.
• Citation Half-Life How long does a newly-cited URL stay cited? Per engine, per content type, per vertical.
• Content Features That Persist Which structural properties (comparison tables, defined acronyms, primary-source citations, schema markup) correlate with sustained citation across engines?
• The Bear Case Where the data is noisy, where engines disagree with each other, where the methodology has limits. We publish what we don’t know.

Early Access Waitlist

First read, lowest price, plus an early-access perk we’re only offering once.

The report will be sold publicly at full price when it ships. Waitlist members get the report first, at a meaningful discount, plus the perks below.