Soda — Declarative Data Quality Monitoring with YAML

Soda offers a YAML-based approach to data quality monitoring. Soda Checks Language lets you define checks without writing code.

Soda Checks Language¶

# checks/orders.yml
checks for fact_orders:
  - row_count > 0
  - missing_count(order_id) = 0
  - duplicate_count(order_id) = 0
  - min(total_czk) >= 0
  - freshness(order_date) < 1d
  - values in (customer_id) must exist in dim_customers (customer_id)

Running Checks¶

from soda.scan import Scan
scan = Scan()
scan.set_data_source_name("warehouse")
scan.add_sodacl_yaml_file("checks/orders.yml")
scan.execute()
if scan.has_check_fails():
    for c in scan.get_checks_fail():
        print(f"✗ {c.name}: {c.outcome}")

Soda vs Great Expectations¶

• Soda — YAML, quick start, SQL-native
• GX — Python API, complex validations

Summary¶

Soda is ideal for a declarative YAML approach to data quality. Quick to start and easy to integrate.