Skip to main content
smart-analyst Smart Analyst uses AI to review conversations from the past 24 hours across all channels—voice, webchat, and SMS. Ask natural-language questions like “What are the top escalation reasons?” and get immediate answers without writing queries. The analyst appears as a chatbot inside Studio, analyzing recent conversations and surfacing insights on agent usage, customer sentiment, and conversation drop-off patterns.

Key benefits

  • Ask freeform questions about conversation data across all channels
  • Identify top containment blockers
  • Surface technical issues, drop-offs, or sentiment trends
  • Track daily agent performance patterns

Deep sampling

Smart Analyst analyzes up to 500 conversations for patterns (10× the previous limit). Conversations are sampled randomly or based on a metric / PolyScore. What you can do:
  • Ask questions like “What do customers complain about?” or “How can we reduce handoffs?”
  • Get directional insights on behavior frequency
  • Uncover patterns that smaller samples miss
  • Validate hypotheses at scale
Deep sampling is useful for exploratory analysis when you want to understand broad trends across many conversations.

Example prompts

Smart Analyst understands natural questions across a range of categories:

Containment & Transfer Insights

  • What are the top 5 reasons conversations are transferred to a human agent?
  • What % of reservations are dropped before confirmation?
  • What are the most common reasons customers don’t complete a reservation?
  • What actions could increase containment for loyalty lookup failures?

Quality & sentiment

  • Do customers express frustration with the assistant?
  • What types of conversations result in negative sentiment?
  • Are customers complaining about pricing or offers?
  • Are customers reporting issues with our mobile app or site?

Conversation analytics

  • What are my top 5 verb–noun pairings across “Speak To” conversations?
  • Which conversations had the longest AHT yesterday?
  • Which intents failed the most often?

Knowledge gaps

  • What kinds of questions are we not handling well?
  • What information does the assistant usually request during booking?
  • How does the assistant handle unavailable dates?

Reservation-specific insights

  • What drives drop-offs in the reservation flow?
  • What are the top blockers during reservation confirmation?

Templates

smart-analyst-template Pre-built templates help you start with ready-made prompts grouped by Customer Insights, Agent Performance, and Custom Research.
  • Each template includes a description and optimized query
  • Selecting a template populates the prompt field — run it or modify as needed
  • Templates help discover insights without structuring queries manually

How It Works

1

Open Smart Analyst

Go to Analytics > Smart Analyst in the sidebar. The interface opens with a chat window and a summary of the latest available conversation data.
2

Ask your first question

Type a natural-language question into the chat, or choose a template to edit. Smart Analyst reviews transcripts and metadata from the previous day to generate a response.
3

Refine and explore

Follow up with related or deeper questions. Smart Analyst keeps context and adjusts its answers based on your ongoing thread.
4

Review and act

Use Smart Analyst to uncover patterns, monitor performance, and prioritise improvements — no need to write SQL.