Skip to main content

Using MetricFlow with dbt metrics

Metricflow now integrates with dbt metrics! Metricflow will parse your dbt project, and translate the result into a Metricflow model. You can then query your metrics in any supported Metricflow interface.

Setting up the integration with MetricFlow​

  1. Configure metrics in your dbt project. For example, we have the metric salesforce_opportunties.
metrics:
  - name: salesforce_opportunities
    label: Opportunities
    model: ref('opportunity_base')
    description: "Count of opportunities"
    calculation_method: count_distinct
    expression: id 
    timestamp: created_date
    time_grains: [day, week, month]
    dimensions:
      - stage_name
    meta: {team: Sales}

  1. To install metricflow for use with dbt you do the following pip install "metricflow[your-dbt-adaptor]" , replacing the your-dbt-adaptor variable with whatever supported data warehouse you use (currently dbt-snowflake, dbt-redshift, dbt-postgres , and dbt-bigquery)

  2. In your .metricflow/confg.yml, the key model_path should point to your dbt project directory. Add the additonal key dbt_repo: "true"

After this is done, all other MetricFlow commands, $ mf command options, should work as normal but now from your dbt metrics! 🎉

For example, you can query the salesforce_opportunities metric with the following command

mf query --metrics salesforce_opportunities --dimensions metric_time