Activity schema data modeling using a single table
summary
Claims that business analytics fit into a single time‑series table with 11 columns (no JSON) where every source row becomes a customer activity (pageview, order, email) and queries use temporal relationships rather than foreign keys. It defines 11 relationship types (example: first-in-between) to link events, shows real migrations (Shopify example) and says transforms average ~25 lines each, runs in production for ~4 years via Narrator, and notes SQL for joins gets hairy but is auto‑generable via a UI.
AI-written summary · about
chapters (8)
- 0:00Introduction to activity schema
- 3:43Problems with traditional data modeling
- 8:36Introducing the activity schema concept
- 12:47Benefits and downsides of activity schema
- 14:32Building an activity schema table
- 17:19Querying the activity schema
- 24:14Key takeaways and future possibilities
- 27:00Q&A
related talks
24mEvent Analytics - from unstructured to insightful
12mAnalytics architecture strategy for intelligent experiences
18mDemystifying Data Modeling
16mGood Data Modeling & Effective Communication in Power BI
18mBehavioral Data Creation for AI - Delivering Value and Insights
57mAlgebraic Databases
34mDimensional Data Modeling in the Modern Era
23mDelivering Retail Analytics & Data Projects With Data Created For Purpose