1. Using swivl
  2. Performance & Analytics

Monitoring Customer Questions (Interactions)

Review and analyze how users engage with your bot

The Interactions section in swivlStudio gives you visibility into every conversation your chatbot has with users. This page helps you understand what people are asking and how effectively the bot is handling those requests.

What You’ll See

  •  A list of recent user queries and the bot’s response to each.
  • Status icons:
    • Handled interaction ✅ The bot successfully understood and responded to the query. 
    • Unhandled interaction ⚠️ The bot didn’t understand the request and couldn’t respond appropriately.
  • The Conversation icon allows you to click through and view the full interaction thread with that user.
  • The Brain column shows which AI model (or “brain”) processed the interaction.

Filtering and Search

Use filters to sort interactions by:

  • Intent – Group messages by what users were trying to do (e.g., pay a bill, check inventory).

  • Handled/Unhandled status

  • Date range

  • Search bar – Find specific messages or keywords.

Once filtered, you can export the data as a CSV for reporting or review.

Intents

Intents reflect the purpose behind a user’s message—what they’re trying to accomplish. For example, messages like “I want to rent a unit” or “What size do I need?” might be tagged with the intents Understand Storage Size or Check Inventory. Reviewing intents helps you identify trends and opportunities to improve your bot’s training and performance.