This is a personal story of a Nordstrom team that inherited a messy relational data store and built a graph replacement from the ground up. It includes the discovery of political headwinds against graph technology based on a previous unsuccessful (and costly) implementation and the way in which this team overcame that opposition. You’ll see an honest breakdown of the process, including both benefits and pitfalls to switching to a graph model. Finally, you’ll discover the tools that were developed and the insights that were gained in the fight against fraud in the e-commerce world of Nordstrom.