Feature overview
Quality Evaluation now supports recurring frequency for conversation evaluations. Previously, only event-based trigger evaluations were supported, meaning evaluations were created only when a conversation closed.
With recurring evaluations, users can run bulk evaluations over previously closed conversations, evaluate future closed conversations, and sample selected conversations that match configured conditions. This enables ongoing quality monitoring, retrospective analysis, and targeted evaluation programs without relying solely on conversation-close events.
Expected release date
October CY2026