CXHub: Voice of Customer (VOC)

Unlock the power of customer feedback with CXHub: Voice of Customer (VOC) to enhance your customer experience strategy and drive loyalty.

Written By Gideon O'Daniel (Administrator)

Updated at July 10th, 2025

Introduction

Welcome to your dedicated Voice of Customer (VOC) platform, designed to provide invaluable insights from customer interactions. This powerful tool captures, analyzes, and helps you understand the voice of your potential and existing customers, enabling you to enhance their experience, refine your offerings, and drive success.

Configuration: Tailoring VOC to Your Business

The Configuration module empowers administrators and analysts to define how VOC tags, filters, and scoring mechanisms work. This includes enabling or disabling AI tagging dimensions, adjusting filters, and creating user roles. Custom taxonomies, multilingual support, and integration points can also be configured here, aligning the platform with your internal analytics frameworks and CRM systems.

 

 

The Dashboard: Your Real-Time Customer Pulse

The Dashboard is your mission control for real-time customer sentiment and interaction trends. It provides at-a-glance insights into what your customers are feeling, asking, and valuing during conversations. Whether you’re tracking daily call volumes or evaluating the performance of a marketing campaign, this module equips decision-makers and managers with the information needed to act quickly.

The Dashboard offers a dynamic and customizable overview of your key VOC metrics. Visualize overall sentiment, understand emerging trends, and quickly identify areas requiring attention – all in real-time.

 

VOC Dashboard Navigation:

  • Login: Access your VOC platform.
  • Navigation: Click on the "Dashboard" tab.
  • Widget Customization: Drag and drop widgets to arrange them as per your preference. Use the "Add Widget" or "Delete Widget" options to tailor the displayed information.
  • Global Filters: Apply filters at the top of the dashboard to analyze data for specific timeframes, campaigns, agents, etc.
  • Widget-Level Interaction: Click on data points within widgets to explore further details or navigate to relevant reports for deeper analysis.

 

VOC Dashboard Key Features:

Customizable Layout: Arrange widgets to prioritize the insights most relevant to your team's focus, whether it's campaign performance, lead quality, or customer satisfaction with specific offerings.

 

 

Real-time Data: Get up-to-the-minute visibility into customer feedback as interactions are processed, allowing for timely interventions and proactive strategies.

Comprehensive Filtering: Apply filters across the entire dashboard based on date range, specific snapshots, sentiment (positive, negative, neutral), individual agents, call dispositions, identified topics, sales campaigns, and agent skills.

 

Pre-configured Widgets: Access standard widgets designed to provide immediate insights.

  • Total Interactions & Duration Processed: 
    • Purpose: Provides a summary of the volume and total length of customer interactions analyzed by the system.
    • Details: Displays the total number of calls and chats processed within the selected timeframe and the cumulative duration of these interactions. This gives you a sense of the overall data being analyzed.
  • Predicted Unsolicited Calls (Spam Detection): 
    • Purpose: To identify and quantify potential unsolicited or spam calls to ensure compliance with regulations and maintain efficient resource allocation.
    • Details: Shows the number of calls flagged by the AI-powered system as likely unsolicited or spam.
       

Custom Widgets (Upon Request): Tailored widgets can be developed to meet your specific analytical needs.

Interactive Widgets: Dive deeper into specific data points by clicking on them to reveal more granular information or navigate to detailed reports.

 

VOC Highlights:

Highlights aspects that customers appreciate and value, directly derived from their positive feedback during conversations in three areas:

  • Details: A bar chart displaying the frequency of positive mentions related to specific strengths (e.g., "knowledgeable agents," "helpful information," "attractive property designs," "efficient booking process").
  • AI Power: LLMs analyze the conversation for expressions of satisfaction, appreciation, and positive attributes related to the organization, its services, or offerings. For instance, phrases like "I was very impressed with...", "The agent was extremely helpful...", or "I really liked the..." are identified and categorized as strengths. The frequency of these positive mentions for different aspects is then aggregated.
  • Usefulness: Understanding your strengths allows you to reinforce what you're doing well, highlight these aspects in marketing, and train new agents on successful communication strategies.

 

Word Cloud: Provides a visual representation of the most frequently used words in customer conversations, offering a quick overview of key themes and vocabulary. This tool displays a cloud of words, with the size of each word indicating its frequency in the analyzed interactions. This can help identify emerging topics or common areas of discussion.

Sales Made Distribution: Tracks the conversion rate of customer interactions into successful sales or desired outcomes. A pie chart showing the percentage of calls or chats that resulted in a "Sales Made" disposition versus those that did not. This provides a high-level view of sales effectiveness through the contact center.

Language Analysis: Studies the primary languages spoken by your customers during interactions, aiding in resource allocation and agent skill development. A pie chart illustrating the distribution of languages spoken in the processed interactions.

 

Sentiment Overview: visualizes the trends of customer sentiment (positive, negative, neutral) over time for specific campaigns, agent skills, or discussed topics.

  • Details: The data is displayed using a line chart. You can select specific sentiment types (positive, negative, neutral) to view. Different lines on the chart can represent different campaigns, skills, or frequently discussed topics. The Y-axis shows the percentage of calls with that specific sentiment.
  • AI Powered: Sentiment in this widget, and throughout the platform, is determined using advanced AI, including Large Language Models (LLMs). These models analyze the nuances of language, including tone, context, specific keywords, and even subtle cues like pauses and changes in speech rate, to classify the overall customer sentiment as positive, negative, or neutral with high accuracy. This goes beyond simple keyword matching, understanding the intent and emotion behind the words.
  • Usefulness: By tracking sentiment trends for specific campaigns or topics, you can understand how customer perception evolves in response to different initiatives or regarding specific aspects of your offerings. Identifying dips in positive sentiment can signal potential issues that need investigation.

 

Hourly Call Volume Trends: identifies peak call times for different campaigns or skills and understand the associated sentiment during those hours.

A table displays campaigns (or agents, skills, or topics) on the left and hourly intervals across the top. Each cell shows the total call volume for that campaign in that hour. By clicking on tabs within the table (Positive, Negative, Neutral), you can see the percentage of calls within that hour for that campaign that expressed that specific sentiment.

This tool identifies hourly trends in call volume and sentiment can help optimize staffing levels and proactively address potential issues that might be arising during specific times of the day.

 

Identifying Opportunities for Call or Agent Improvement

Identifies areas where customers express dissatisfaction or encounter issues, providing crucial insights for improvement:

  • Details: A bar chart illustrating the frequency of negative mentions associated with specific pain points (e.g., "pricing concerns," "lack of information on specific amenities," "difficult scheduling process," "delays in response").
  • AI Power: LLMs analyze the conversation for expressions of dissatisfaction, frustration, complaints, and negative feedback regarding specific aspects. Phrases like "I was disappointed with...", "This is unacceptable...", "I'm having trouble with...", or questions indicating confusion or unmet expectations are identified and categorized as pain points.
  • Usefulness: Identifying pain points is critical for addressing customer concerns, improving processes, and preventing future dissatisfaction, ultimately leading to increased customer retention.

 

VOC Reports: In-Depth Analysis of Customer Interactions

Reports allow you to zoom into each conversation with precision. This module supports your team in identifying recurring customer concerns, evaluating advisor performance, and extracting specific feedback from customer interactions that may not be immediately visible in dashboard trends.

The Reports module provides comprehensive and customizable analyses of your customer interactions, allowing you to delve deep into the data and uncover specific insights that drive strategic decisions.

 

VOC Reports: Key Features

 

Detailed Interaction Report: A comprehensive table containing every captured detail of each customer interaction.

Customizable Columns: Choose which data points are displayed in your reports based on your specific analytical needs.

Advanced Filtering: Apply granular filters to isolate specific sets of interactions based on various criteria (same as dashboard filters).

Data Export: Export report data in multiple formats (CSV, Excel) for offline analysis and integration with other business intelligence tools.

 

VOC Reports: Detailed Interaction Report

 

This report provides a granular view of each customer interaction, with the following columns:

  • UCI (Unique Call Identification): A unique identifier for each individual customer interaction (call or chat).
  • Channel: Indicates the communication channel used for the interaction (e.g., Voice, Chat).
  • Interaction Type: Specifies the type of outbound or inbound call (e.g., Predictive, Progressive, Review, Manual, Inbound). For chat, this might indicate the initiation method.
  • Start Time: The exact date and time when the customer interaction began.
  • Agent Name: The name of the agent who handled the interaction.
  • Topics: The key subjects and themes discussed during the interaction, automatically identified by our AI-powered topic analysis system.
    • AI Power: Our AI, leveraging LLMs, analyzes the conversation transcript to identify the core topics discussed. It goes beyond simple keyword matching to understand the context and identify nuanced themes. For example, if a customer mentions "the balcony view is amazing" and then asks about "maintenance charges," the topics identified might be "Property Feature - View" (classified as positive) and "Fees - Maintenance."
    • Usefulness: Understanding frequently discussed topics helps you identify customer interests, common questions, and potential areas of confusion.
    • Campaign/Skill: The name of the marketing campaign or skill-based routing queue.
  • Sentiment: The overall sentiment expressed by the customer during the interaction (Positive, Negative, or Neutral).
    • AI Power: As with the dashboard, sentiment analysis in the reports is powered by advanced AI, including LLMs. The system analyzes the entire conversation transcript, considering tone, word choice, and context to determine the overall emotional state of the customer.
    • Usefulness: Identifying calls with negative sentiment can trigger follow-up actions, while positive sentiment can highlight successful interactions.
  • Call Date: The date on which the customer interaction occurred.
  • Time (Duration): The total length of the customer interaction.
  • Hold Time: The total duration for which the customer was placed on hold.
  • Agent Comment: Any notes or comments entered by the agent.
  • Call Disposition: The outcome or status assigned to the call by the agent.
  • Interaction Summary: A concise, AI-generated summary of the entire customer interaction.
    • AI Power: LLMs are used to automatically generate these summaries, extracting the most important information from the conversation transcript.
    • Usefulness: Quickly understand the gist of a customer interaction without reading the entire transcript.
  • Interaction Script: Clicking on the "View Script" button opens a pop-up with the complete conversation transcript.

 

VOC Reports: Interaction Transcript Pop-Up

 

This pop-up provides a detailed view of a single customer interaction.

  • Transcript: The entire conversation, segmented by speaker (Agent and Customer).
  • Summary: The AI-generated summary of the conversation.
  • Sentiment: The overall customer sentiment for this interaction.
  • Topics: The specific topics identified within this interaction.
  • All Interaction Report Details: All other data points from the main Interaction Report.
  • Play Recording: Option to play the audio recording of the call (if enabled).
  • Transcription Verification: Allows you to verify the accuracy of the AI-powered transcription.

 

VOC Reports: Generating and Using Reports:

Navigation: Click on the "Reports" tab.

Report Selection: Choose the "Detailed Interaction Report."

Column Selection: Use the "Customize Columns" option to select the data points you want to include.

Apply Filters: Utilize the comprehensive filtering options to narrow down the data.

View Data: The filtered data will be displayed in a tabular format.

Export Data: Click the "Export" button to export the report data.

View Interaction Script: Click the "View Script" button to access the transcript pop-up.