Case study · Snowflake, dbt, and Tableau
One governed model for multi-brand ecommerce operations
A multi-brand ecommerce portfolio had sales, finance, inventory, refunds, fulfillment, marketing, and creator data spread across systems with conflicting business logic. Greyfield rebuilt reporting in Snowflake and dbt with shared rules for brands, products, fees, refunds, dates, and cancellations.
to answer a cross-brand question
dbt models behind it
records reconciled
Snowflake
dbt
Fivetran
Airbyte
TableauAt a glance
The challenge
The portfolio needed comparable answers across brands and channels, but source systems represented products, fees, refunds, dates, and cancellations differently. Those disagreements had to be resolved before a dashboard could be trusted.
Approach
What Greyfield did
Organize seven operating domains
Organized commerce, finance, inventory, fulfillment, advertising, web, and creator data into staging, intermediate, and reporting layers in Snowflake and dbt.
Resolve cross-platform logic
Made brand, product, date, fee, cancellation, and refund rules explicit. During one Shopify connector migration, 53 tests passed and 10,105 overlapping order-line records reconciled before release.
Build shared Tableau facts and dimensions
Built facts and dimensions in dbt so Tableau workbooks could use the same calculations instead of rebuilding them inside each workbook.
Results
What changed for the client
Greyfield organized seven operating domains into 120 dbt models. Tableau now reads shared facts and dimensions across brands. Before connector changes reach production, tests compare record counts and overlapping order lines; one Shopify migration reconciled 10,105 records after 53 tests passed.
to answer a cross-brand question
dbt models behind it
records reconciled
Have a report or data workflow like this?
Discuss a similar project