Interaction Analytics Metrics

• Calculation: The total number of distinct cases where the client frustration indicator equals 1. This metric represents whether any instance of client frustration occurred during the selected period.

• Filters:

  • Employee group: Agent, Team, Company

  • Contact group: Company

  • Attributes: NA

• Supported channel: All Channels

• Metric type: Historical

• Metric direction: Down, a lower metric value is best.

• Use case: Identifying Customer Frustration During Interactions In a contact center, detecting moments of customer frustration is critical for improving service quality and customer satisfaction. This metric helps identify whether any interaction during the reporting period involved a frustrated customer (where the system flagged Client Frustration ID = 1). For example, if a customer repeatedly expresses dissatisfaction during a call or chat—such as complaining about delays or failed transactions—the interaction is marked as frustrated. Tracking this metric allows supervisors to quickly spot sessions that may require review, agent coaching, or escalation.

  By monitoring frustration indicators, contact centers can:

  • Pinpoint areas where processes or policies cause customer dissatisfaction.

  • Take proactive steps to resolve recurring issues.

  • Improve agent training for handling difficult conversations.

  • Enhance overall customer experience by reducing frustration triggers.

• Calculation: This metric indicates whether any interaction was successfully resolved during the selected period.

• Filters:

  • Employee group: Agent, Team, Company

  • Contact group: Company

  • Attributes: NA

• Supported channel: All Channels

• Metric type: Historical

• Metric direction: Up, a higher metric value is best.

• Use case: Tracking this metric allows supervisors to quickly see if resolutions are happening during the reporting period. A low value signals that no issues were resolved, which may indicate a backlog or process inefficiency. By monitoring this, contact centers can improve resolution rates, reduce repeat contacts, and enhance customer satisfaction.

• Calculation: The total number of distinct interaction segments within the selected period. Each segment represents a portion of an interaction, such as a call, chat, or other communication channel activity.

• Filters:

  • Employee group: Agent, Team, Company

  • Contact group: Company

  • Attributes: NA

• Supported channel: All Channels

• Metric type: Historical

• Metric direction: Up, a higher metric value is best.

• Use case: In a contact center, a single customer call might involve multiple segments—such as the initial conversation with an agent, a transfer to a specialist, and a follow-up confirmation. Similarly, a chat interaction could include separate segments for different topics or escalations. Counting distinct segments provides insight into how many interaction parts agents manage, which can indicate complexity and resource needs.

• Calculation: The average duration of interaction segments during the selected period. This metric is calculated by dividing the sum of all distinct segment durations by the number of distinct segments. It represents the typical length of a segment within an interaction.

• Filters:

  • Employee group: Agent, Team, Company

  • Contact group: Company

  • Attributes: NA

• Supported channel: All Channels

• Metric type: Historical

• Metric direction: Down, a lower metric value is best.

• Use case: In a contact center, a single customer call might include multiple segments—such as the initial greeting, troubleshooting, and escalation. If the total duration of these segments is 900 seconds and there are 3 segments, the average segment duration is 300 seconds (5 minutes). Monitoring this metric helps managers identify patterns like unusually long troubleshooting steps or short, rushed interactions, enabling better resource planning and training.

• Calculation: The average duration of interaction segments during the selected period. This metric is calculated by dividing the sum of all distinct segment durations by the number of distinct segments. It represents the typical length of a segment within an interaction.

• Filters:

  • Employee group: Agent, Team, Company

  • Contact group: Company

  • Attributes: NA

• Supported channel: All Channels

• Metric type: Historical

• Metric direction: Down, a lower metric value is best.

• Use case: In a contact center, a call might include troubleshooting steps where the agent is searching for information or waiting for system responses, resulting in silence. If the total silence across segments is 120 seconds and there are 4 segments, the average silence per segment is 30 seconds. High silence averages may signal the need for better agent training, faster tools, or improved processes to keep customers engaged.

Learn more

• Calculation: The proportion of interaction segments where the Client Sentiment ID equals negative sentiments. This metric is calculated by dividing the number of distinct segments with sentiment ID equals negative sentiments by the total number of segments. It represents the percentage of segments that reflect a specific sentiment.

• Filters:

  • Employee group: Agent, Team, Company

  • Contact group: Company

  • Attributes: NA

• Supported channel: All Channels

• Metric type: Historical

• Metric direction: Down, a lower metric value is best.

• Use case: In a contact center, if sentiment analysis marks segments where customers express satisfaction as Negative Sentiment, this metric calculates what portion of all segments were negative. For example, if there are 100 segments and 40 of them have Negative Sentiment, the metric will show 40%. Monitoring this helps managers track trends in customer sentiment, identify areas for improvement, and evaluate the impact of service changes on customer perception.

• Calculation: The percentage of interaction segments where Positive Client Sentiment. This metric is calculated by dividing the number of distinct segments with Positive sentiments by the total number of segments. It represents the proportion of segments that reflect a specific sentiment

• Filters:

  • Employee group: Agent, Team, Company

  • Contact group: Company

  • Attributes: NA

• Supported channel: All Channels

• Metric type: Historical

• Metric direction: Up, a higher metric value is best.

• Use case: In a contact center, if sentiment analysis marks segments where customers express dissatisfaction as Positive Sentiment, this metric calculates the share of all segments that were positive. For example, if there are 200 segments and 50 of them have Positive Sentiment, the metric will show 25%. Monitoring this helps managers detect trends in positive sentiment, take corrective actions, and improve overall customer experience.

• Calculation: The percentage of interaction segments where Agent Sentiment Negative Sentiment. This metric is calculated by dividing the number of distinct segments with negative agent sentiment by the total number of segments. It represents how often agents display negative sentiment during interactions.

• Filters:

  • Employee group: Agent, Team, Company

  • Contact group: Company

  • Attributes: NA

• Supported channel: All Channels

• Metric type: Historical

• Metric direction: Down, a lower metric value is best.

• Use case: In a contact center, sentiment analysis might detect negative agent sentiment when an agent uses harsh language, sounds impatient, or expresses frustration during a call or chat. For example, if there are 100 total segments and 8 segments show negative sentiment (Agent Sentiment), the metric will display 8%. Tracking this helps supervisors spot patterns of negative sentiment, provide coaching, and ensure agents maintain professionalism and empathy during interactions.

• Calculation: The percentage of interaction segments Positive Agent Sentiment. This is calculated by dividing the number of distinct segments flagged with Positive Agent Sentiment by the total number of segments.

• Filters:

  • Employee group: Agent, Team, Company

  • Contact group: Company

  • Attributes: NA

• Supported channel: All Channels

• Metric type: Historical

• Metric direction: Up, a higher metric value is best.

• Use case: In a contact center, if sentiment analysis marks segments where customers express satisfaction as Positive Sentiment, this metric calculates the share of all segments that were positive. For example, if there are 200 segments and 50 of them have Positive Sentiment, the metric will show 25%. Monitoring this helps managers detect trends in positive sentiment, take corrective actions, and improve overall customer experience.

• Calculation: The percentage of interaction segments where the client began the interaction with negative sentiment. This is calculated by dividing the number of distinct segments where Negative Client Begin Sentiment by the total number of segments.

• Filters:

  • Employee group: Agent, Team, Company

  • Contact group: Company

  • Attributes: NA

• Supported channel: All Channels

• Metric type: Historical

• Metric direction: Down, a lower metric value is best.

• Use case: In a contact center, if sentiment analysis detects that a customer begins a call or chat sounding upset or frustrated (Negative Client Begin Sentiment), this metric calculates the proportion of such segments. For example, if there are 100 total segments and 15 started with negative sentiment, the metric will show 15%. Monitoring this helps managers understand customer sentiment trends at the start of interactions and take steps to improve first impressions—such as reducing hold times or enhancing self-service options.

Learn more

• Calculation: The percentage of interaction segments where the client began the interaction with a positive sentiment. This is calculated by dividing the number of distinct segments where Positive Client Begin Sentiment by the total number of segments.

• Filters:

  • Employee group: Agent, Team, Company

  • Contact group: Company

  • Attributes: NA

• Supported channel: All Channels

• Metric type: Historical

• Metric direction: Up, a higher metric value is best.

• Use case: In a contact center, if sentiment analysis detects that a customer begins a call or chat sounding happy or satisfied (Positive Client Begin Sentiment), this metric calculates the proportion of such segments. For example, if there are 100 total segments and 40 started with positive sentiment, the metric will show 40%. Monitoring this helps managers understand customer mood at the start of interactions and take steps to maintain or improve positive engagement.

• Calculation: The percentage of interaction segments where the client ended the interaction with a negative sentiment. This is calculated by dividing the number of distinct segments where Negative Client End Sentiment by the total number of segments.

• Filters:

  • Employee group: Agent, Team, Company

  • Contact group: Company

  • Attributes: NA

• Supported channel: All Channels

• Metric type: Historical

• Metric direction: Down, a lower metric value is best.

• Use case: In a contact center, if sentiment analysis detects that a customer ends a call or chat sounding frustrated or dissatisfied (Negative Client End Sentiment), this metric calculates the proportion of such segments. For example, if there are 100 total segments and 20 ended with negative sentiment, the metric will show 20%. Monitoring this helps managers identify interactions that failed to meet customer expectations and take corrective actions to improve resolution and customer experience.

• Calculation: The percentage of interaction segments where the client ended the interaction with a positive sentiment. This is calculated by dividing the number of distinct segments where Positive Client End Sentiment by the total number of segments.

• Filters:

  • Employee group: Agent, Team, Company

  • Contact group: Company

  • Attributes: NA

• Supported channel: All Channels

• Metric type: Historical

• Metric direction: Up, a higher metric value is best.

• Use case: In a contact center, if sentiment analysis detects that a customer ends a call or chat sounding happy or satisfied (Positive Client End Sentiment), this metric calculates the proportion of such segments. For example, if there are 100 total segments and 60 ended with positive sentiment, the metric will show 60%. Monitoring this helps managers evaluate service quality, identify best practices, and improve processes to ensure customers leave interactions feeling positive.

• Calculation: The percentage of interaction segments where the agent began the interaction with a negative sentiment. This is calculated by dividing the number of distinct segments where Negative Agent Begin Sentiment by the total number of segments.

• Filters:

  • Employee group: Agent, Team, Company

  • Contact group: Company

  • Attributes: NA

• Supported channel: All Channels

• Metric type: Historical

• Metric direction: Down, a lower metric value is best.

• Use case: In a contact center, sentiment analysis might detect negative agent sentiment at the start of a call if the agent sounds impatient, frustrated, or uses harsh language. For example, if there are 100 total segments and 10 started with negative sentiment, the metric will show 10%. Monitoring this helps supervisors spot patterns of negative tone at the beginning of interactions and take corrective actions to improve professionalism and customer experience.

• Calculation: The percentage of interaction segments where the agent began the interaction with a positive sentiment. This is calculated by dividing the number of distinct segments where Positive Agent Begin Sentiment by the total number of segments.

• Filters:

  • Employee group: Agent, Team, Company

  • Contact group: Company

  • Attributes: NA

• Supported channel: All Channels

• Metric type: Historical

• Metric direction: Up, a higher metric value is best.

• Use case: In a contact center, sentiment analysis might detect positive agent sentiment at the start of a call if the agent greets the customer warmly and sounds enthusiastic. For example, if there are 100 total segments and 70 started with positive sentiment, the metric will show 70%. Monitoring this helps supervisors ensure agents consistently begin interactions on a positive note, which can lead to better outcomes and higher customer satisfaction.

• Calculation: The percentage of interaction segments where the agent ended the interaction with a negative sentiment. This is calculated by dividing the number of distinct segments where Negative Agent End Sentiment by the total number of segments.

• Filters:

  • Employee group: Agent, Team, Company

  • Contact group: Company

  • Attributes: NA

• Supported channel: All Channels

• Metric type: Historical

• Metric direction: Down, a lower metric value is best.

• Use case: In a contact center, sentiment analysis might detect negative agent sentiment at the end of a call if the agent sounds impatient, uses harsh language, or expresses frustration before ending the conversation. For example, if there are 100 total segments and 12 ended with negative sentiment, the metric will show 12%. Monitoring this helps supervisors spot patterns of negative tone at the close of interactions and take corrective actions to improve customer experience and agent performance.

• Calculation: The percentage of interaction segments where the agent ended the interaction with a positive sentiment. This is calculated by dividing the number of distinct segments where Positive Agent End Sentiment by the total number of segments.

• Filters:

  • Employee group: Agent, Team, Company

  • Contact group: Company

  • Attributes: NA

• Supported channel: All Channels

• Metric type: Historical

• Metric direction: Up, a higher metric value is best.

• Use case: In a contact center, sentiment analysis might detect positive agent sentiment at the end of a call if the agent thanks the customer warmly and offers additional help before ending the conversation. For example, if there are 100 total segments and 75 ended with positive sentiment, the metric will show 75%. Monitoring this helps supervisors ensure agents consistently maintain a positive tone throughout the interaction, especially at closing, which leaves a lasting impression on customers.

• Calculation: The total sum of all distinct client sentiment segments recorded during the selected period. Each segment ID represents a unique interaction segment where client sentiment was analyzed. This metric essentially aggregates the unique sentiment-related segments to provide an overall count of sentiment occurrences.

• Filters:

  • Employee group: Agent, Team, Company

  • Contact group: Company

  • Attributes: NA

• Supported channel: All Channels

• Metric type: Historical

• Metric direction: Up, a higher metric value is best.

• Use case: In a contact center, a single call might include multiple segments—such as greeting, troubleshooting, and closing—each analyzed for sentiment. If sentiment analysis identifies 3 unique segments in one call and 2 in another, this metric sums those distinct segment IDs across all interactions. Monitoring this helps managers understand how frequently sentiment analysis is applied and how many segments contribute to overall sentiment trends.

• Calculation: The total sum of all distinct agent sentiment segments recorded during the selected period. Each segment ID represents a unique interaction segment where the agent’s sentiment was analyzed. This metric aggregates the unique sentiment-related segments to provide an overall count of segments associated with agent sentiment.

• Filters:

  • Employee group: Agent, Team, Company

  • Contact group: Company

  • Attributes: NA

• Supported channel: All Channels

• Metric type: Historical

• Metric direction: Up, a higher metric value is best.

• Use case: In a contact center, a single call might include multiple segments—such as greeting, troubleshooting, and closing—each analyzed for agent sentiment. If sentiment analysis identifies 3 unique segments in one call and 2 in another, this metric sums those distinct segment IDs across all interactions. Monitoring this helps managers understand how frequently agent sentiment analysis is applied and how many segments contribute to overall sentiment trends.