Copilot best practices

[Image]

In this article, we'll guide you through the functionalities of our AI-powered copilot system, its restrictions, and how you can maximize its potential to benefit your workflows.

⚠️ [Admin role] Requirements / How to start

• Go to your Organization page

• Organization Name

• Organization description

Good practices

• Look for specific terms that are close to your context. Avoid generic search criteria.
  For example, a bank, does not look for Improvements on “Cards“ but rather “Credit or Debit Cards“

• Before taking action based on Copilot’s recommendations, check the feedback used to make the recommendations.

What’s coming?

• Feb 2024

  • Copilot working with Filters

  • Saving the recommendations from Feedier Copilot

  • General query

RAG (Retrieval Augmented Generation), our latest AI enhancement, leverages advanced technology to analyze and generate insights from the vast customer feedback stored. It can perform specific tasks such as identifying popular topics, listing improvements, pinpointing pain points, summarising feedback, and providing an overall sentiment score.

How RAG Works:

Currently, the following tasks are available:

1. Return the List of Main Topics:

   • Provides a list of the topics most frequently mentioned in the customer feedback related to your search criteria.

2. List the Top Improvements:

   • Presents the top suggested enhancements mentioned in customer feedback related to your specified search item.

3. List the Main Positive Points:

   • Identifies and lists the most positive points expressed in customer feedback regarding your search criteria.

4. Synthesis:

   • Generates a concise summary of the customer feedback related to your specified search item.

How to Use RAG Effectively:

To extract the maximum value from RAG, follow these steps:

1. Specify Task Type:

   • Choose the appropriate task type from the dropdown provided based on your analysis needs.
     

     

2. Enter Search Criteria:

   • Define the search item to narrow down the scope of customer feedback for more targeted results.

   • Example “Delivery”
     

     

3. Generate your AI result:

   • Select the Generate button.
     

     

4. Understanding Feedier AI response.
   

   

5. Iterate for Deeper Analysis:

   • Repeat the process for different tasks or search items to gain a comprehensive understanding of customer sentiments and preferences.

Restrictions:

While RAG is a powerful tool, it's essential to be aware of its limitations:

• Data Scope:

  • Results are based on the data available in our Elastic search database. Ensure your search criteria align with the stored customer feedback in Feedier platform.

• Task Specificity:

  • RAG excels at the specified tasks mentioned above but may not perform optimally for tasks beyond its designed scope.

• Continuous Improvement:

  • Our AI system is continually evolving. Expect periodic updates to enhance its capabilities and address any limitations.