Many financial institutions allow customers to contact them to ask questions and resolve issues. During these conversations, customers may offer unique insights into their lives, and indirectly, their financial needs. We present an industry case study on the use of Large Language Models (LLMs) to identify significant events in customers' lives from conversational data. Through identification of these events, financial institutions can offer increased personalization during interactions or with promotions. We use Gemini Pro 1.5 to extract these events, with preliminary results demonstrating the effectiveness of our approach. We validate the LLM’s performance on a conversational dataset from a leading financial institution. Beyond initial results, we provide key insights and future directions for others interested in applying similar techniques with their own conversational data.