Introduction

Citeroot records what AI answer engines say when your buyers ask questions, traces each answer to its sources, and checks your site for what keeps you out of those answers.

#What it does

1. Collects real answers. You write (or generate) a panel of buyer questions. Citeroot asks each one of ChatGPT, Claude, Gemini and Perplexity, with web search switched on, and stores the full answer and every source it cited.
2. Measures them. For each answer it records whether your brand was named (and in what order), whether your domain was cited, whether the answer reads as a recommendation, and which competitors were named.
3. Traces sources. Source Map shows which domains the answers lean on, and which cite your competitors but never you.
4. Audits your site. Site Audit checks AI-crawler access in robots.txt and scores each page for answer-readiness.
5. Drafts the next step. Briefs and Insights (beta) turn those findings into a draft you review.

#What it doesn't do

• It doesn't publish anything, contact anyone or change your site.
• It doesn't track Google AI Overviews, Copilot, Meta AI or Grok.
• It isn't a replica of a logged-in consumer chat. See Engines and models.
• Analytics and Search Console integrations, an API, webhooks, an MCP server and team seats are planned, not built. See Exports and roadmap.

#How the docs are organised

• Getting started — this page, the quickstart and the concepts.
• Measuring AI visibility — metric definitions, prompt panels, sources, variance, engines and the monthly answer allowance.
• Diagnosing your site — Site Audit, crawler access, briefs and insights.
• Connect — what you can export today and what is planned.
• Account — trial, billing, security and your data.