Conversation Analyzer - customer sentiment

1 Overview
- 1.1 Sentiment scores
2 Filters
3 Tiles

Overview

The Conversation Analyzer – customer sentiment dashboard provides insights into customer and agent sentiment during conversations. It includes analytics for inbound and outbound interactions across both voice and digital channels.

You can use filters to refine the data. When you apply a filter, the dashboard automatically refreshes to display data for the selected criteria.

Sentiment scores

Sentiment	Sentiment score range	Color

Sentiment	Sentiment score range	Color
Very negative	–1 <= score < –0.6	Dark red
Negative	–0.6 <= score < –0.2	Red
Neutral	–0.2 <= score < 0.2	Yellow
Positive	0.2 <= score < 0.6	Green
Very positive	0.6 <= score < 1	Dark green

Filters

You can set the following global filters for this dashboard:

Name	Description	Default setting

Name	Description	Default setting
Date	The period of time you want to analyze data for. You can choose different preset ranges, or you can specify your own custom data range.	Last 7 days
Profile	The profile (set of categories, subcategories, and rules) you want to analyze data for.	All profiles
Category	The category you want to analyze data for.	All categories
Subcategory	The subcategories you want to analyze data for.	All subcategories
Disposition code	The outcomes of the conversations you want to analyze data for.	All disposition codes
Initial direction	The direction of the conversation when it started. Available options are inbound, outbound, and internal.	All directions
Media type	The communication channel through which the conversation occurred. Available options are phone, email, SMS, WhatsApp, and webchat.	All media types
Agent	The agent(s) you want to analyze data for, identified by name.	All agents
Group	The group(s) you want to analyze data for, identified by name. You can filter agents assigned to the selected group or groups.	All groups and all agents
Skill	The skill you want to analyze data for, identified by name.	All skills and all agents
First queue	The first queue you want to analyze data for, identified by name. The first queue is where the conversation was initially routed to.	All queues
Subcategory matches	The number of times the subcategory was matched during the conversation.	A value greater than or equal to 0
Interaction transferred	Whether the conversations you want to analyze data for have been transferred. Available options are Yes and No.	All conversations
Recording time	The total time of the conversation recording. Depending on the configuration, this might include time spent in call flow applets and queues. The Duration unit setting determines whether this value is displayed in milliseconds, seconds, minutes, hours, or days.	All time
Duration unit	The duration unit for the recording time you want to analyze. Available options are: Milliseconds Seconds Minutes Hours Days	Minutes
Customer hold time (ms)	The average time an agent put the conversation on hold in milliseconds.	All time
Overall sentiment score	The overall sentiment scores you want to analyze data for.	Is not null.

Tiles

Name	Description	Data included

Name	Description	Data included
Average sentiment score	Displays the average sentiment score calculated across all conversations.	Average sentiment score
Average sentiment score trends	Displays how the average of all conversations’ average sentiment scores changes over time.	Average sentiment score The date the conversations occurred
Conversations by sentiment type	Displays the number of conversations that were, overall, either very positive, positive, neutral, negative, or very negative.	Sentiment type
Conversations by sentiment type trends	Displays the daily count of conversations categorized as very positive, positive, neutral, negative, or very negative.	Sentiment type The date the conversation occurred
Conversations by sentiment type by agent	Displays the number of conversations handled by the agent, grouped by their overall sentiment: very positive, positive, neutral, negative, or very negative.	Sentiment type Agent name
Sentiment distribution by agent	Breaks down the agent’s every conversation to show the average percentage of conversation spent in each sentiment category: very positive, positive, neutral, negative, or very negative. This differs from the Conversations by sentiment type by agent tile, which only shows the overall sentiment per conversation.	Sentiment type Agent name
Average sentiment score by agent trends	Displays the average sentiment of all conversations handled by the agent.	Average sentiment score The date the conversation occurred
Average sentiment trends (by category)	Displays how the average sentiment score changes over time by category. This does not mean that a specific category has a specific sentiment.	Category names Average sentiment score The date the conversation occurred
Average sentiment trends (by sub category)	Displays how the average sentiment score changes over time by subcategory. This does not mean that a specific subcategory has a specific sentiment.	Subcategory names Average sentiment score The date the conversation occurred
Declining sentiment by agent trends	Displays the number of times a conversation started poorly (first 30% of the conversation) and declined towards the end (last 30% of the conversation).	Category names Stacked subcategory names Number of times the subcategory was found
Improving sentiment by agent trends	Displays the number of times a conversation started poorly (first 30% of the call) and improved towards the end (last 30% of the call).	Category names Stacked subcategory names Number of times the subcategory was found
Conversation details	Displays basic details about the conversation, such as start time, initial direction, and subcategories found.	N/A

Conversation details tile columns

Name	Description	Example value

Name	Description	Example value
Conversation ID	Unique identifier for the conversation.	c5aa8aa7-4e2d-4e8a-8d76-3d229d7274ae
Start time	The date and time at which the conversation started.	2026-02-15 09:32:14
Name	The agent’s name.	JSmith
Overall sentiment	A verbal representation of the overall sentiment score.	Positive
Overall sentiment score	An average of sentiment scores throughout the call.	0.72
Sentiment delta	A value calculated by comparing the average sentiment of the first 30% of the conversation against the final 70%. If the difference is: less than 0, then Sentiment delta is Declined; equal to 0, then Sentiment delta is No Change; greater than 0, then the Sentiment delta is Improved;	No change
Handle time	The time an agent has spent working on a conversation. Displayed in hh:mm:ss format.	00:15:42
Talk time	The time the customer and agent spent in the connected channel state. This column excludes any time the customer was on hold. Displayed in hh:mm:ss format.	00:12:18
Recording time	The total of the recorded time of the conversation. This can be longer than the talk and handle time, but possibly less than the connected time. Displayed in hh:mm:ss format.	00:14:05
Customer hold time	The average time the agent put the conversation on hold. Displayed in hh:mm:ss format.	00:02:30
Very negative sentiment (%)	The percentage of the conversation that is considered very negative.	5%
Negative sentiment (%)	The percentage of the conversation that is considered negative.	10%
Neutral sentiment (%)	The percentage of the conversation that is considered neutral.	15%
Positive sentiment (%)	The percentage of the conversation that is considered positive.	50%
Very positive sentiment (%)	The percentage of the conversation that is considered very positive.	20%
Sentiment score for … of conversation	The average sentiment for the conversation, divided into ten 10% sections: 0-10% 10-20% … 80-90% 90%-100%	-0.6
Media type	The communication channel through which the conversation occurred. Available options are phone, email, SMS, WhatsApp, and webchat.	Phone