Discovery
Discovery answers the question your analytics can't: what does your category look like to AI models, and where are you missing? It runs structured research sessions that turn AI conversations into findings you can act on.
acme.comRun research from a template#
Each template is a packaged research method (a Category Visibility Check, a Market Opportunity scan) that directs your AI through a multi-step session instead of one shallow conversation.
Templates unlock as your Business Profile fills in: Market Opportunity, for example, needs your category phrase to know what to research. Locked templates stay visible and tell you exactly which profile detail unlocks them.
Discovery runs in conversational AI surfaces (claude.ai or Claude Desktop) where long research sessions are natural. The dashboard is where sessions are launched from and where results land.
Review sessions over time#
Every session is kept with its status and structured output. Skim summaries in the session list, or open one for the full per-query responses when you want to see the evidence behind a finding.
Turn findings into keyword clusters#
Research that ends as a document is research that gets forgotten. After a session, derive keyword clusters straight from the findings, pre-grouped with live volume data, and they land in Keywords ready to drive content.
From your AI tool
Everything above is also reachable over MCP: same data, no dashboard tab. The toolbar's MCP handoff on this page pre-fills the project context for you.
Discovery also has a family of agent-invoked bookkeeping tools (discovery_submit_*) the AI calls mid-session; you never invoke them directly. They're documented under MCP Tools.