Avoiding Data Disillusionment: 3 Things to Get Right When Building Data Products

Businesses are pouring massive amounts of money into data science projects, and expectations are sky-high. But how many of those projects will deliver real value to customers? The history of other hyped new technologies predicts that many will fail, leaving a sense of disillusionment in their wake. Can the data science community do better?

In this talk, James Psota will share three key lessons learned in the process of building Panjiva over the last decade—lessons that can help reduce the risk of failure with data products. The learnings will draw from the Panjiva story, spanning from idea to profitable business to a successful exit. The talk will also share a brief overview of Panjiva, which illuminates the massive and complex global supply chain using machine learning and text analytics to extract insights from over a billion transaction-level shipping records. Panjiva has spent years learning how to turn messy data into intelligence using data science, scaling to thousands of customers. Join this timely discussion about building data products that deliver real value.