What is it?
The Pathlight CI Sentiment feature analysis evaluates the emotional tone of a conversation. We offer two types of sentiment analysis: Agent Sentiment and Customer Sentiment. This analysis method aids in determining the feelings expressed by the customer and the agent during their interaction. Pathlight categorizes sentiment into three categories: Positive, Neutral, and Negative. By understanding these sentiments, businesses can better respond to customer needs and improve communication efficacy.
How does it work?
The Large Language Model (LLM) analyzes the words and phrases used by the customer and the agent to decide if their sentiment is positive, neutral, or negative. Since the LLM looks at the sentiment across the entire conversation, it accurately assesses it while removing any momentary aspects of sarcasm, irony, etc. Each conversation receives insights into both Agent Sentiment and Customer Sentiment. We can use these insights in aggregate to understand agent performance and customer satisfaction. It's important to highlight that the model evaluates the entire conversation comprehensively and won't be misled by isolated words or expressions.
Benefits of the Feature:
Understanding Agent Sentiment is important to ensure that Agents always provide high-quality customer support. This feature saves time as it does it automatically without the need to listen or read the transcript.
Naturally, it is also important to understand if your customers are happy with your service. Customer Sentiment is a far more effective indication of customer satisfaction than a post-interaction survey - which the customer will often ignore.
We hope this was helpful! Please submit a ticket here if you have any questions or need further assistance.