...
Info |
---|
Exploration → |
...
Assignments |
...
In this article, we
...
present the features of our Eureka AI
...
assignment system and how you can
...
maximise its potential
...
by giving it assignments to carry out through the use of filters and prompts.
Feedier's assignment uses real-time
...
data to provide you with the best possible answers.
...
It is important to note that the
...
assignment is only designed
...
for specific tasks.
Return
...
list of main topics:
Provides a list of the
...
most frequently mentioned topics in
...
customer
...
reviews related to your search criteria.
List
...
of main improvements:
Presents the
...
main improvements suggested
...
in customer feedback
...
in relation to
...
the specified search item.
...
Main
...
positives list:
Identifies and lists the most positive points expressed in customer feedback
...
relating to your search criteria.
...
Summary:
Generates a concise summary of
...
customer feedback
...
relating to
...
the specified search item.
...
⚠️ [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:
Return the List of Main Topics:
Provides a list of the topics most frequently mentioned in the customer feedback related to your search criteria.
List the Top Improvements:
Presents the top suggested enhancements mentioned in customer feedback related to your specified search item.
List the Main Positive Points:
Identifies and lists the most positive points expressed in customer feedback regarding your search criteria.
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:
...
Specify Task Type:
Choose the appropriate task type from the dropdown provided based on your analysis needs.
...
Enter Search Criteria:
Define the search item to narrow down the scope of customer feedback for more targeted results.
Example “Delivery”
...
Generate your AI result:
Select the Generate button.
Understanding Feedier AI response.
...
...
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:
...
Table of Contents | ||||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|---|
|
Before you start using Assignment [Administrators Only]
Info |
---|
Profile → Organization |
In order to get the most accurate results to your organisation, it’s important to input the following to support the prompt on Assignment:
Add the name of your organisation (if not added already)
Add principal category of feedback collected
Add a clear and detailed description of your organisation.
Understanding the available actions within the Assignment
Task drop-down list: In this drop-down list, you can choose between the 4 tasks: Main improvements, Main positive points, synthesis, and Main topics.
Search prompt entry: In this search field, you can enter text to make the results more precise, or leave it blank.
Global Filter: Here you can set filters to funnel the feedback you wish to see eg. survey x, or time period after 01/10/24.
Saved Assignments: You can bookmark your searches, and they will appear on the left hand side of the screen.
Results: A summary, followed by key themes with attached pertinent feedback can be found here.
Attributes: Based on the summary generated, attributes that are strongly linked are marked in green, generally linked in grey, and loosely linked in red.
Running through an example
Select the desired task → for example: Main improvements
Enter a prompt → for example: Customer Service
Choose the filters you wish to apply to the search → for example: Time period equals last 30 days.
Click Generate.
How do you save an assignment?
Click on the bookmark icon to the right to save the assignment. Once saved, it will appear in the list of saved assignments. Saved assignments are visible only to you.
How can you retrigger the same assignment you did before?
You can always revisit your triggered assignments by clicking on the date and selecting an earlier date.
Pre filled Reports based on an Assignment
Our new feature allows you to create a pre-filled report based on a Eureka AI assignment in one click, selecting the "Create Report" icon in the Assignment page will do it. The report will include the assignment result on the first page and a pre-filled section with all related feedbacks.
This ensures comprehensive and relevant data representation.
Steps to create the report:
After selecting the icon, the Feedier Report modal creation opens:
Eureka AI Assignment Component
Introducing the Assignment Component in Feedier Report: seamlessly integrate Eureka AI assignments into Feedier Reports. This feature lets you display and manage saved assignments and results directly in your reports. Initially, it will show saved assignments and results, allowing for a streamlined and organised reporting process. Future iterations will include options to add new results and generate results from new requests with global filters.é
How it works:
Have a saved assignment in Assignment Page
Create a Report
Create a Eureka AI Assignment Component:
Component Creation:
Select Your Saved Assignment
Choose the Data Version: Display the desired result version.
Title Configuration: Default title is the prompt + task, but customisable.
Displayed Title
Applied Filters: View filters applied to the assignment.
Data Version: Displayed result version.
Additional tips:
Search for specific terms close to your context. Avoid generic search criteria. For example, a bank might not be looking for improvements on "cards" but rather on "credit or debit cards".
Before acting on Assignment's recommendations, check the information used to formulate the recommendations.