Analytics | Advanced Sentiment dashboard

The Analytics | Advanced Sentiment dashboard provides a split-channel sentiment analysis, separating customer (user) sentiment from agent sentiment across all transcribed contacts. It enables business units to understand not only how users felt during interactions but also how agents were presenting — two distinct and complementary views of interaction quality. Topic-level sentiment breakdowns and a phrase-level word cloud provide additional context for identifying the drivers of positive and negative sentiment.

This dashboard is designed for Quality Managers, Operations Managers, and CX Leaders who want a split-channel view of sentiment to understand both the customer experience and agent behaviour on the same interactions.

Questions this dashboard helps answer:

What percentage of contacts have positive, neutral, and negative customer sentiment?
What percentage of contacts have positive, neutral, and negative agent sentiment?
What topics are most associated with negative customer sentiment?
What phrases are most commonly found in positive and negative customer sentiment contacts?
How are customer and agent sentiment trending over time?
Which specific contacts had notable sentiment for deeper review?

Prerequisites

You have the Insights Reader license.
You must have access to Insights dashboards with permission to view Analytics data.
You must have access to the Contacts and Phrase Usage datasets.

Page location

Insights > Dashboards > Analytics | Advanced Sentiment

Dashboard layout

The dashboard contains the following sheets:

Advanced Sentiment Analysis

The primary sheet. Contains side-by-side customer and agent sentiment KPI panels, customer and agent sentiment trend charts, topic tables for positive and negative customer sentiment, and phrase word clouds.
Interaction Drill Down

A contact-level detail table with Overall Agent Sentiment and Overall Customer Sentiment filters, allowing drill-down to specific contacts matching the selected sentiment combination.

Controls and filters

NOTE Filters are hierarchical. Selecting a Group limits the available Teams and Agents. Clicking KPIs or charts may automatically apply sentiment-focused filters and navigate to the drill-down sheet.

From / To (date pickers) — both sheets

Sets the date range for all sentiment data.
Group(s) / Team(s) / Agent(s) (dropdowns) — both sheets

Standard hierarchical scope filters.
Overall Agent Sentiment / Overall Customer Sentiment (list filters) — Drill Down sheet only

Filter the contact drill-down table to contacts with a specific agent or customer sentiment classification. Use to isolate contacts where, for example, customer sentiment was negative but agent sentiment was positive — a potential indicator of an agent handling a difficult interaction well.

Procedures

Compare customer and agent sentiment

Set the date range and apply scope filters.
Review the customer and agent sentiment KPI panels side by side.
Review both trend charts to assess whether customer and agent sentiment are moving together or diverging.
Review the topic tables to identify the primary drivers of negative customer sentiment.

Use this to:

Identify contacts where agent sentiment was negative despite positive customer outcomes — or vice versa.
Use topic and phrase data to build targeted coaching materials.

Drill into specific contacts by sentiment combination

Navigate to the Interaction Drill Down sheet.
Apply the Overall Customer Sentiment and/or Overall Agent Sentiment filters to select the combination of interest.
Review the contact list and click contact URLs to open specific records.

Use this to:

Find specific contacts where agent and customer sentiment diverge for calibration or coaching review.

Key metrics

Metric	Definition	How to use it
Sheet: Advanced Sentiment Analysis
Customer Sentiment KPIs (% Positive / Neutral / Negative)	Percentages of contacts classified as positive, neutral, and negative for customer sentiment. Based on the overall customer (user) sentiment channel.	The primary CX health metrics. Rising negative customer sentiment is the most direct signal of deteriorating customer experience.
Agent Sentiment KPIs (% Positive / Neutral / Negative)	Percentages of contacts classified as positive, neutral, and negative for agent sentiment. Based on the overall agent sentiment channel.	Agent sentiment reflects tone and language on the agent’s side. High agent negative sentiment may indicate stress, disengagement, or handling challenges. Use alongside customer sentiment for a complete interaction picture.
Contacts with Sentiment	Total count of contacts for which sentiment analysis has been completed.	Use to assess coverage. If significantly lower than total contacts, some contacts are not being transcribed or sentiment analysis is not enabled for all queues.
Customer Sentiment Trend (bar chart)	Trend chart showing positive, neutral, and negative customer sentiment contact counts over time.	Track whether customer sentiment is improving or declining, and identify specific weeks where sentiment shifted.
Agent Sentiment Trend (bar chart)	Trend chart showing positive, neutral, and negative agent sentiment contact counts over time.	Use alongside the customer sentiment trend. If agent negative sentiment rises before customer negative sentiment, agent behaviour changes may be a leading indicator of future CX degradation.
Topics by Negative Customer Sentiment (table)	A table showing topics associated with negative customer sentiment contacts, with counts and percentages.	Identify the top topic drivers of negative customer sentiment. Use to focus root cause analysis and coaching on the most impactful areas.
Topics by Positive Customer Sentiment (table)	A table showing topics associated with positive customer sentiment contacts.	Identify what drives positive outcomes. Use to reinforce training and coaching around topics that consistently produce satisfied customers.
Phrase Word Clouds	Word clouds for phrases found in negative and positive customer sentiment contacts, sized by frequency.	Surface recurring language patterns. High-frequency negative phrases are priority candidates for coaching and process review. It is common to see the same phrases in interactions with positive sentiment and negative sentiment.

Data notes

Data is sourced from the Contacts and Phrase Usage datasets.
Twelve calculated fields are used to compute sentiment percentages for both agent and customer channels separately.
Split-channel sentiment requires both agent and customer audio channels to be transcribed. Single-channel transcription will only produce overall sentiment data.