This page defines all metrics and scores used in AIdapt, both system-generated and custom.
These metrics are calculated automatically by AIdapt for every imported conversation.
A numerical score generated by AI that reflects how well the AI agent handled the conversation. The score is accompanied by a short comment explaining the reasoning behind it.
Use the Quality Score to quickly identify poorly handled conversations and prioritize them for review.
A concise, AI-generated summary of the conversation in a few sentences. Useful for understanding what a conversation was about without reading the full transcript.
A more comprehensive, structured AI-generated summary that breaks down the conversation into key aspects such as the customer's intent, the agent's response, and the outcome.
Indicates whether the conversation was escalated from the AI agent to a human agent. AIdapt detects handovers automatically from the conversation data.
Handover Rate is the aggregate KPI showing what percentage of all conversations resulted in a handover.
The percentage of conversations that were fully resolved by the AI agent without any human involvement. A higher automation rate generally indicates a better-performing agent.
A classification of how long and complex a conversation was, based on the number of messages, duration, and topic complexity. Classified as short, medium, or long.