AI Improved Operational Efficiency in Customer Service Environments

Summary:

Generative AI has revolutionized the processing of unstructured data, turning it into actionable insights, which helps enhance decision-making and boosts operational efficiency in the call center.

Client: 

Financial services provider

Problem Statement: 

A financial services provider faced a major challenge: their call center generated large volumes of unstructured data through call transcripts. While they had some metadata and tags from customer service representatives (CSRs), these were often too general and lacked the specificity needed to generate actionable insights. 

The client wanted a deeper understanding of the calls, including topics discussed, questions raised, issues encountered, and the complexity of interactions. This information was essential for improving customer service, training, and overall operational efficiency.

The main objective was to leverage an AI tool to extract valuable insights from these untapped sources of unstructured data (call transcripts) and convert them into structured data assets.

Results: 

• Significantly reduced number of calls to CSRs.
• Boosted operational efficiency.
• Enhanced customer satisfaction.
• Created structured data from unstructured call transcripts.

AI Solution Overview:

Fulcrum extracted a detailed list of topics discussed during each call, providing insights at the individual call level and enabling the aggregation of topics into broader categories. They combined AI processing with the client’s pre-existing topic categories, which involved several iterations of processing, human review, and different approaches to mapping topics to categories.

This mapping process required both AI and human expertise to ensure accuracy and relevance. They tested various methods, including machine learning models and ad hoc mapping, to create a set of categories that were both comprehensive and manageable for reporting.

The final extracted data points were integrated into a complete data asset, viewable in Excel. This asset included the original call transcript, metadata, and all the extracted data, such as topics discussed, their categories, FAQs, complexity ratings, issues or complaints, and actions taken.

References: 

1. Case Study: Transforming Call Center Insights with Generative AI. https://fulcrumanalytics.com/2024/09/10/case-study-transforming-call-center-insights-with-generative-ai/

Industry: Human Resources

Vendor: Fulcrum

Client: Financial services provider

Publication Date: 2024