How to get Snowplow data models to do more for you
Blog post from Snowplow
Over the past few years, the snowplow-web dbt package has evolved significantly, offering enhancements that allow users to derive more value from their data with increased ease and speed. The latest updates focus on reducing the need for custom models by making it possible to add fields from the atomic events table directly to derived tables, such as page_views, sessions, and users, simply by setting variables. This streamlines the process that previously required creating custom models for each new field. The package introduces customizable session identifiers and a set of base macros that encapsulate core sessionization logic, providing users with the flexibility to tailor the package's functionality to their specific needs. These improvements not only enhance usability but also optimize performance, as seen in the ability to use load_tstamp for efficient session timestamping. The package aims to handle complex sessionization scenarios through manifest tables, ensuring complete and accurate processing of session data while minimizing costs. Looking ahead, the developers plan to extend these features to additional packages and explore new integrations to unify web and mobile data.