How it Works

The expect_column_quantile_values_to_be_between test from the dbt-expectations package validates that a specified quantile of a column's values falls within a defined range. This enables flexible distribution checks without relying on mean or standard deviation, which can be skewed by outliers.

The quantile argument specifies the quantile to evaluate (e.g., 0.5 for the median, 0.95 for the 95th percentile).

Steps and Conditions

1. Column Selection: Choose the numeric column.

2. Quantile: Specify the quantile value between 0 and 1.

3. Define Range: Set min_value and/or max_value.

4. Execution: The specified quantile is computed and compared to the range.

5. Outcome: Pass if within range; fail otherwise.

Example Usage: Performance Monitoring

A platform engineering team monitors API response times and wants to ensure the 95th percentile latency stays below 500ms.

models:
  - name: api_request_logs
    columns:
      - name: response_time_ms
        tests:
          - dbt_expectations.expect_column_quantile_values_to_be_between:
              quantile: 0.95
              min_value: 0
              max_value: 500

models:
  - name: api_request_logs
    columns:
      - name: response_time_ms
        tests:
          - dbt_expectations.expect_column_quantile_values_to_be_between:
              quantile: 0.95
              min_value: 0
              max_value: 500

models:
  - name: api_request_logs
    columns:
      - name: response_time_ms
        tests:
          - dbt_expectations.expect_column_quantile_values_to_be_between:
              quantile: 0.95
              min_value: 0
              max_value: 500

This ensures that the vast majority of API requests are served within performance SLA bounds.