The workflow before
44 Farms ran its business on a Ruby on Rails application hosted on Heroku. A single Postgres database handled customer transactions and every analytics query. Heavy reports competed with live orders for the same database, which put the store's performance at risk and capped how much analysis the team could run.
What we built
We designed a decoupled data pipeline that replicates data from the production Postgres database into a dedicated cloud data warehouse. Reporting and analysis now run against the warehouse, not the database that takes orders.
Airflow orchestrates the replication and modeling jobs, and dbt models turn raw tables into clean, tested reporting tables. Pipelines are built to fail safely and rerun without manual cleanup.
Two Blue Orange data engineers worked inside the 44 Farms engineering team, shipping alongside their developers and putting standard practices in place: version-controlled dbt models and CI/CD for data changes.
Before and after
| Measure | Before | After |
|---|---|---|
| Where reporting runs | Production Postgres database | Dedicated cloud warehouse |
| Reporting load on the store's database | Every report and ad hoc query | None |
| Data modeling | Ad hoc queries in the app database | Version-controlled dbt models with CI/CD |
What's next
With clean, modeled data in a warehouse, 44 Farms has the foundation for Level 2 (Connected Intelligence): an analytics assistant that answers sales and inventory questions from the warehouse in plain English.
