cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 
4995members
1526posts
This is a verified product documentation article. For case-based resolutions articles, please reference the Knowledge Base section of Invoca Community.
Jacaselon
Community Manager
Community Manager

Contents of this Article

Introduction

Sentiment Analysis in Invoca automatically detects and highlights moments of positive and negative sentiment directly within call transcripts, and provides an overall caller sentiment score for each call. This valuable data is readily accessible through reports, dashboard tiles, and can even trigger Smart Alerts via Signals.

This capability is designed for all Invoca users, including contact center leaders, marketers, and multi-location businesses focused on enhancing customer experiences.

Note: You must have the Signal Gold or QM package to enable this feature. For more information about obtaining this feature, reach out to your CSM or success@invoca.com
.

Benefits and Use Cases

Invoca's AI-powered Sentiment Analysis goes beyond basic call metrics, identifying the overall sentiment of every conversation. This means you can quickly pinpoint positive and negative sentiment, zero in on key call highlights, and uncover valuable opportunities for improvement – to enhance your customer experiences and drive better outcomes.

Here’s how you can leverage Sentiment Analysis in our platform to enhance your experience.

Invoca Platform

Use Cases

Invoca for Contact Center

High/Low Sentiment Identification:

Pinpoint calls for targeted agent coaching and feedback.

CSAT/DSAT Trend Analysis:

View sentiment scores by region, agent, or other specified details., to analyze satisfaction trends.

Specific Call Review:

Easily locate and review call moments with positive or negative sentiment for CX insights.

Rules-Based Signals / Sentiment Alerts:

Get notified of defined sentiment trends.

Invoca for Marketing

Sentiment Analysis:

Identify positive, neutral, and negative sentiment in reports.

Correlation/Trend Detection:

Analyze sentiment scores to find relationships and patterns with Signals

Regional/Location Aggregation:

Gain insights by analyzing sentiment trends across different areas.

Rules-Based Signals / Sentiment Alerts:

Set up notifications for specific sentiment trends.

 

Sentiment Analysis in Calls Reports

  1. Login to your Invoca portal and select the Reports icon in the left menu bar, and then select Calls located under Detail Reports.

    Sentiment Analysis 14.png

  2. You can locate Sentiment scores in the AI Summary tab as shown here.

    Sentiment Analysis 15.png

  3. Select the multi-squares icon on the top left of the Calls Report window, and the Call Details report and diagrams will appear.

  4. Scroll to the right until you locate the Sentiment and Sentiment Score columns.
    Sentiment Analysis 16.png


Sentiment Analysis in the Call Review Console

Note: Although Sentiment is available for all Signal AI Gold and QM customers, you must have the QM package to enable this view.

  1. Login to your Invoca portal and select the Call Review icon in the left menu bar, and your calls will appear.

  2. You can use the Filters in the Call Review window to filter to your specific calls you’re looking for such as the Date Range, Scores, Evaluated By, Agent, Review Status, Signals, Global Transcript Search, Caller ID, Comments, Marketing Data, Sentiment, and More Filters.

    Sentiment Analysis 06.png

  3. In the Sentiment Filter you can filter exclusively to Positive, Negative, or Neutral Sentiments as shown here.

    Sentiment Analysis 02.png

  4. You can additionally Filter your calls from Highest to the Lowest Sentiment scores, as well as Most to Least Recent using the dropdown filter in the upper right corner.

    Sentiment Analysis 04.png

  5. When you click on a particular call, you’ll then be able to view the Sentiment results throughout the call along with a Sentiment score.

    Sentiment Analysis 05.png

 

Sentiment Analysis in the Signal Builder

  1. Log in to your Invoca account, and then click Signal located in the left sidebar menu.

  2. Select Manage Signals.

    Sentiment Analysis 17.png

  3. Locate the Signal you’d like to add Sentiment Analysis to, and click to open it.

    Note: You can find the Signal by using the search bar or scrolling through the list. If you’d like to create a New Signal, visit this resource to learn more.

    Sentiment Analysis 18.png

  4. The Call Details section is located on the left side of the window, scroll down until you locate Sentiment and then click it.

  5. The Sentiment signal will be added to the Signals menu. You can Select Sentiment : Equals or Not Equals : Positive, Neutral, or Negative.

    Sentiment Analysis 07.png

  6. Click Update and the changes will be saved and applied to the Signal.

Smart Alerts for Sentiments

  1. Log in to your Invoca account, select the bell icon located in the top right corner.

    Sentiment Analysis 08.png

  2. Click Add Smart Alert.

    Sentiment Analysis 20.png

  3. In the Smart Alerts window, select the blue +Add Alert button.

    Sentiment Analysis 09.png

  4. Enter the Name you’d like to use for this alert into the Name field as shown here.
    Sentiment Analysis 21.png

  5. Select the start and end times you want the alerts to be monitored during in the When We Monitor section by clicking on the clock icons.

    Sentiment Analysis 11.png
  6. Select the desired Signal using the dropdown arrow, and determine whether you’d like the Signal to be True, False, or Not Applied.

  7. Choose the time window you want the alert to apply to.

  8. Then determine the threshold number of calls by either the actual number OR percentage of calls.

    Sentiment Analysis 13.png
  9. Enter the emails of the people you’d like to send the alerts to in the Recipients field, and click Add Users to save.

    Sentiment Analysis 22.png

  10. Select Save and your Smart Alerts will be set!
Need more help?

Don't see what you are looking for? You can ask the Community or contact support.