Yellow.ai announced the launch of Analyze. Powered by an in-house LLM model, Analyze reduces ticket volume by 30% and boosts containment rates by 10%.
Traditional automation platforms provide limited insights, focusing only on basic metrics like user numbers or session times. This gap leaves businesses lacking a comprehensive understanding of chatbot interaction quality. According to a recent Yellow.ai survey, 54.5% of customer service professionals seek to enhance their data analysis capabilities through AI adoption. They are turning to AI-first solutions to gain comprehensive insights into bot effectiveness, user satisfaction, conversation topics, and opportunities for improvement in bot interactions.
Addressing this demand, Yellow.ai’s Analyze not only delivers detailed insights but also uses this information to continuously improve the bot’s ability to handle a broader range of customer queries without human intervention.
“Customer interactions and contact centre data hold immense potential to elevate customer experience, yet many businesses are missing out due to outdated technology,” said Raghu Ravinutala, CEO & Co-founder of Yellow.ai. “With the launch of Analyze, we aim to meet this market need and help enterprises close gaps in their customer service strategies. Analyze provides comprehensive metrics that enhance containment opportunities and drive more effective automation.”
Analyze accomplishes this through four key features:
- Next-Generation Self-Learning Loopback Technology: Analyze’s self-learning functionality enhances automation for voice and chat bots. When a customer query is escalated to a human agent, the transcript is fed back into the system to generate knowledge base articles. These articles enrich the company’s knowledge base, enabling the bot to handle similar conversations more effectively in the future.
- Strategic Insights for Topic Clustering: It enables customer service teams to explore AI-generated topic clusters from bot conversations through an intuitive interface. They can access topic-wise insights on customer sentiments, potential knowledge base article improvements, conversation share, and containment rate opportunities.
- Conversation Analysis for Improved Customer Support: It analyses customer conversations to improve the quality of resolution and customer satisfaction. With Analyze, teams can access granular, conversation-level reports instantly, allowing them to assess details such as, resolution status, containment rate opportunity, conversation share and more.
- Sentiment Analysis for Higher User Satisfaction: Using deep learning, Analyze categorises conversations as positive, negative, or neutral, offering deeper insights into resolution quality. This analysis, applied to topic clusters, provides more reliable data compared to traditional self-reported feedback.
“Insights into bot and user conversations are crucial for us. Analyze by Yellow.ai has the potential to be transformative with in-depth conversation intelligence. The Self-Learning Loopback using LLMs to study human agent conversations, create KB articles and enhance bot automation, stands out. We are excited to see how this can help drive high quality customer service automation,” said Eric Hansen, Chief Information Officer, Waste Connections.
“This solution evolves with the business, becoming increasingly powerful and adept at meeting customer needs with each interaction,” said Ravinutala. “We believe it represents a breakthrough in customer service analytics, giving businesses a significant edge to maximise their ROI from AI-first automation.”
