AI Tagging & Sentiment

Automatic topic tags and sentiment scores on every call and chat session.

How it works

After each call or chat session completes, the platform automatically analyzes the transcript. Two things are extracted:

  • Tags — up to 5 topic tags per conversation (for example: "billing", "appointment", "complaint", "pricing"). Tags are shown as chips on each row in the call log.
  • Sentiment — each conversation is scored Positive, Neutral, or Negative based on the overall tone of the interaction. Sentiment is shown as a colored badge in the log.
Tip · No configuration required. Tagging and sentiment analysis run automatically for all agents on all channels — voice calls, website chat, and SMS.

Where to see it

Go to /dashboard/logs. Each row in the call log shows the extracted tags as small chips and the sentiment as a badge (green for Positive, grey for Neutral, red for Negative).

Open a specific call to see the full tag list and the sentiment score alongside the transcript.

Filtering by tag or sentiment

Use the filter controls at the top of the Logs page to narrow the list to a specific sentiment value or tag. Filtering lets you quickly surface patterns — for example, all calls tagged "complaint" with Negative sentiment from the past week.