Conversation AI: New Query Detection Flow

Modified on Tue, 10 Sep at 2:30 PM



What's New:

We are excited to introduce a significant upgrade to Conversation AI: the Enhanced Training Flow. Previously, query analysis depended on the suggestion flow and the LLM's tools functionality, which was sometimes inaccurate. This new feature analyzes recent conversations and adapts to your sub-account's unique data, ensuring more accurate and context-aware responses.

Key Highlights:

  • Conversation Analysis: The flow reviews the last 5 messages of conversation on each AI request to identify if the contact needs support or has business-related queries like pricing, services, etc.

  • Dynamic Query Extraction: Automatically extracts accurate queries from conversations and provides an analysis score.

  • Bot Training Utilization: Uses specific bot training data from your sub-account to give tailored responses to the identified queries.

How it Works:

  1. Automated Analysis: Each conversation is reviewed to understand the context and identify the contact's exact query.

  2. Query Analysis Score: Provides an analysis score and reasons whether support is needed or if there is a customer query.

  3. Adaptive Responses: Uses the identified query to apply the most relevant training data from your sub-account's bot, ensuring precise and relevant responses.


This update is based on continuous feedback and collaboration with our valued customers.

Note: 

No action is required from users. These changes have been applied internally to enhance the efficiency and effectiveness of AI responses.


Was this article helpful?

That’s Great!

Thank you for your feedback

Sorry! We couldn't be helpful

Thank you for your feedback

Let us know how can we improve this article!

Select at least one of the reasons

Feedback sent

We appreciate your effort and will try to fix the article