Re-thinking the structure of event data

Snowplow emphasizes the importance of structuring event data to ensure data meaning, quality, and governance, which are crucial for deriving business value. Event data captures actions performed by users or services, and it can be collected in various ways, such as unstructured JSON or fixed-schema formats. Snowplow advocates for using self-describing, versioned schemas stored in a schema repository called Iglu, which allows for validation and organized storage in data warehouses. This approach offers the flexibility of unstructured data and the predictability of fixed schemas, enabling data to be descriptive and consistently understood over time. It also allows for the enforcement of data requirements and adaptability to changing data capture needs through semantic versioning. Ultimately, this method supports becoming a data-mature company by ensuring that decisions are informed by reliable data that is well-understood and trusted across the organization.