Demystifying Data Modeling
summary
Example pipeline reads three CSVs, computes per-customer aggregates, and writes a dimensional customer model used to email lapsed customers — the emails reengage buyers. Uses dbt-style practices and a toy Jaffle Shop project, enforces tests like customer_id UNIQUE NOT NULL and loaded_at < 36 hours, and applies a write-audit-publish deploy pattern. High-value definition converges to lifetime value >= $30 plus >3 orders, and documentation plus tests keep metrics debuggable and reproducible.
AI-written summary · about
chapters (10)
- 0:00Introduction to data modeling
- 1:11Defining the business problem
- 2:13Executing the data pipeline
- 4:26Analyzing results and workflow
- 5:34Documentation and testing
- 8:44Deployment and quality control
- 10:15Organizational standardization
- 12:21The value of data modeling
- 13:52Scaling data systems
- 15:09Getting started and final tips
related talks
34mDimensional Data Modeling in the Modern Era
29mUnified Customer Data Management
43mScaling Data Stacks: How To Build a Great Data Model
55mData Modeling for Software Engineers
01h 10mReal World Iterative Database Development
32mMaking Models Work: How an optimized ML lifecycle can drive better ROI
42mData Modelling: Navigating the Data Seas with a Data Vault Compass
29mHow to build a data pipeline without data