video1.0<iframe src="https://www.loom.com/embed/0c6be95a59fc4d38b247f010253e90d0" frameborder="0" width="1114" height="835" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>8351114Loomhttps://www.loom.com8351114https://cdn.loom.com/sessions/thumbnails/0c6be95a59fc4d38b247f010253e90d0-f1662e4422dba48f.gif464.112This Loom explains the Customer Concierge Report metrics for evaluating AI automation performance. It distinguishes engagement rate, which measures how many conversations involved the AI agent, from escalation rate, which measures how often the AI transferred the conversation to a human rep, and AI resolution rate, which measures how many conversations were resolved end to end without a human sending a message. The report breaks down these metrics by date and channel so dips in chat do not get hidden in averages. It also compares customer satisfaction between AI-automated conversations and human-only conversations day by day and points viewers to performance tables by automation, agent, procedure, and tools to identify what is escalating, failing, or erroring out.