Case Study: Vanguard boosts customer support accuracy and speed with Pinecone

How Vanguard worked with Pinecone to boost customer support with faster calls and 12% more accurate responses

Vanguard, a leading investment management company, faced a challenge with its customer support representatives, who were relying on a slow, inefficient keyword-based search to find answers in lengthy financial documents. This process increased call times and reduced customer satisfaction, especially during peak seasons. Seeking a more accurate and real-time retrieval solution, Vanguard’s team chose to implement a RAG-powered system using the Pinecone vector database.

The solution involved building an AI assistant called Agent Assist, powered by Pinecone's hybrid retrieval capabilities. This combined semantic and keyword search to query financial documents. The implementation resulted in 12% more accurate search results, significantly faster call times, and the ability to handle peak periods without adding operational overhead. Pinecone also enabled enhanced compliance through advanced metadata filtering.

Vanguard

Hung Pham

ML Engineer