How it Works

The expect_table_row_count_to_equal_other_table_times_factor test from the dbt-expectations package validates that the row count of one model equals the row count of a reference model multiplied by a constant factor. This is particularly useful for models that fan out from a parent (e.g., generating one row per metric per entity).

Steps and Conditions

1. Reference Model: Specify the comparison model using compare_model.

2. Factor: Define the multiplication factor.

3. Execution: Both row counts are computed; the reference count is multiplied by the factor and compared.

4. Outcome: Pass if counts match; fail if they differ.

Example Usage: Metric Fanout

A data team generates one row per metric per user in a user_metrics model, with 5 metrics per user. The users model has the base row count.

models:
  - name: user_metrics
    tests:
      - dbt_expectations.expect_table_row_count_to_equal_other_table_times_factor:
          compare_model: ref('users')
          factor: 5

models:
  - name: user_metrics
    tests:
      - dbt_expectations.expect_table_row_count_to_equal_other_table_times_factor:
          compare_model: ref('users')
          factor: 5

models:
  - name: user_metrics
    tests:
      - dbt_expectations.expect_table_row_count_to_equal_other_table_times_factor:
          compare_model: ref('users')
          factor: 5

If any user is missing some metrics or extra metrics are generated, the row count will deviate and the test will fail.