Presented by Teresa DeVore, Customer Success Leader, FactorLab

Field observations are one of the most powerful defensive systems for identifying hazards, preventing incidents, and strengthening safety culture. Yet many safety professionals face two persistent challenges: capturing meaningful safety data from field and evaluating the quality of those observations at scale. This session explores how artificial intelligence and Natural Language Processing can transform everyday safety observations into actionable insight into system health. Attendees will learn a practical framework for evaluating observation quality, helping turn raw field data into predictive indicators that go beyond traditional lagging metrics. Through real-world examples and compelling visualizations drawn from millions of work hours, the discussion will demonstrate how observation data can reveal emerging risks, highlight effective leaders, uncover operational constraints, and track improvement over time.

The session will also candidly address lessons learned during implementation, including strategies for increasing engagement when adoption slows. Participants will leave with practical ideas for using data-driven insights to strengthen safety culture and support operational performance.