dbt Mesh Platform

Project Overview

This is a dbt mesh platform project for Jaffle Shop, an e-commerce analytics data platform built with dbt and BigQuery. The project implements a staging layer that cleanses and standardizes raw e-commerce data from various sources.

Data Architecture

Data Sources

• E-commerce Data (ecom source): Raw transactional data from jaffle_shop_raw schema
  • Customer data
  • Order transactions
  • Product catalog
  • Store locations
  • Order line items
  • Supply chain data

Models Structure

Staging Models (models/staging/)

All staging models are materialized as views and follow consistent naming conventions:

• stg_customers: Standardized customer data with cleaned IDs and names
• stg_orders: Order-level data with normalized totals, tax amounts, and timestamps
• stg_order_items: Individual line items linking orders to products
• stg_products: Product catalog with pricing, descriptions, and categorization (food/beverage items)
• stg_locations: Store location data
• stg_supplies: Supply chain and inventory data

Data Governance

The project uses dbt groups for data ownership:

• Platform Team: Jaff Bridges ([email protected])
• Finance Team: Zach Jaff ([email protected])
• Marketing Team: Ben Jaffleck ([email protected])

Data Quality

• Freshness checks configured for critical tables (orders, stores)
• Consistent field naming and type standardization
• Price normalization (cents to dollars conversion)
• Boolean flags for product categorization

Advanced Features