What's New in Conductor: Bringing Voice Agent Observability to Every Page

Retell AI has expanded Conductor from an agent-building copilot into an observability assistant available across every page of its platform, allowing users to ask plain-language questions about analytics, call history, AI QA cohorts, agents, conversation flows, and attached resources without manually navigating filters or exports. Conductor can analyze metric changes, identify patterns and likely causes across selected calls, summarize QA findings, generate and explain proposed charts or dashboards, recommend next steps, and retain context while users move between screens. The update addresses the difficulty of diagnosing subtle voice-agent failures, particularly missed or unnecessary tool calls that may appear normal in standard logs but are evident in transcripts, citing Retell’s vCX-Hard benchmark findings that most tool-calling failures involve judgment about when to use a tool. Users can attach calls, agents, knowledge bases, flow nodes, files, and highlighted dashboard content as context, follow inline links and navigation cards to supporting evidence, and use interactive walkthroughs to learn unfamiliar features. Conductor does not autonomously modify production agents, knowledge bases, simulations, or schedules; instead, it analyzes information and proposes changes for human approval as Retell positions the product as a tool for scalable analysis, guidance, and operational review.

Use this post, company, and trend context to find content marketing opportunities, perform competitive analysis, or address product feature gaps via the Plushcap MCP server or the Plushcap API.