7.3. GDPR Compliance with dbt

Overview

This guide shows a portable approach to marking dbt models and columns as GDPR-relevant and applying tags at the warehouse level automatically. The pattern uses dbt meta on models/columns plus a simple post-hook macro that runs ALTER TABLE/ALTER COLUMN statements. Adjust SQL syntax to match your warehouse (Databricks Unity Catalog, Snowflake, BigQuery, etc.).

What you implement

A model-level opt-in flag to mark a table as GDPR-relevant.
Optional column-level tags for sensitive fields (for example, user identifiers).
A post-hook macro that sets/unsets warehouse tags based on meta.

Setup

Add macros to your project (example files in this repository):

Macro: set_gdpr_compliance_tag

-- Generic GDPR tagging macro for dbt
-- Notes:
-- - Example SQL below uses Databricks/Unity Catalog syntax for tags.
-- - If your warehouse uses a different syntax (e.g., Snowflake), adjust the ALTER statements accordingly.

{% macro set_gdpr_compliance_tag() %}
    {% if execute %}
        {# Table-level meta flag to opt-in the model for GDPR tagging #}
        {% set is_gdpr_model = model.meta.get('is_gdpr_model') %}

        {% if is_gdpr_model %}
            {# Table-level tag #}
            {% set table_tag_name = 'gdpr_deletion' %}
            {% set table_tag_value = 'enabled' %}

            {% set table_level_tag_sql %}
                ALTER TABLE {{ this }} SET TAGS ('{{ table_tag_name }}' = '{{ table_tag_value }}');
            {% endset %}
            {% do run_query(table_level_tag_sql) %}

            {# Column-level tags: set gdpr_column_type when provided #}
            {% set column_tag_name = 'gdpr_column_type' %}
            {% for column in model.columns %}
                {% set col_meta = model.columns.get(column).meta or {} %}

                {# Support either a boolean flag or explicit type string #}
                {% set explicit_type = col_meta.get('gdpr_column_type') %}
                {% set is_flagged = col_meta.get('is_gdpr_column') %}

                {% if explicit_type is string and explicit_type|length > 0 %}
                    {% set column_level_tag_sql %}
                        ALTER TABLE {{ this }}
                        ALTER COLUMN {{ column }} SET TAGS ('{{ column_tag_name }}' = '{{ explicit_type }}');
                    {% endset %}
                    {% do run_query(column_level_tag_sql) %}
                {% elif is_flagged %}
                    {% set column_level_tag_sql_default %}
                        ALTER TABLE {{ this }}
                        ALTER COLUMN {{ column }} SET TAGS ('{{ column_tag_name }}' = 'user_identifier');
                    {% endset %}
                    {% do run_query(column_level_tag_sql_default) %}
                {% else %}
                    {% set column_level_untag_sql %}
                        ALTER TABLE {{ this }}
                        ALTER COLUMN {{ column }} UNSET TAGS ('{{ column_tag_name }}');
                    {% endset %}
                    {% do run_query(column_level_untag_sql) %}
                {% endif %}
            {% endfor %}
        {% endif %}
    {% endif %}
{% endmacro %}

In your dbt model YAML, annotate tables/columns via meta:

Example model YAML with GDPR meta

version: 2

models:
  - name: user_events_monthly
    description: Monthly user engagement events used for activity analytics.
    meta:
      is_gdpr_model: true
      retention: 400d
    columns:
      - name: user_id
        description: Anonymized user identifier
        meta:
          gdpr_column_type: user_identifier
      - name: session_id
        description: Analytics session identifier
        meta:
          is_gdpr_column: true
      - name: event_date
        description: Logical partition/event date
      - name: event_type
        description: Type of event

  - name: advertising_campaigns
    description: Campaigns table retained permanently for financial audit.
    meta:

Configure a post-hook on the models where you want the macro to run. You can set this at the folder level (models: in dbt_project.yml) or per-model using config(post_hook=...).

Usage

Model-level: set meta: { is_gdpr_model: true }. The macro will set a table tag gdpr_deletion=enabled.
Column-level: either set meta: { is_gdpr_column: true } to mark a field as sensitive with default type user_identifier, or provide an explicit type via meta: { gdpr_column_type: <type> }.
The macro also unsets gdpr_column_type on columns that aren’t flagged, keeping tags consistent when schemas evolve.

Warehouse syntax notes

Databricks/Unity Catalog: uses ALTER TABLE <table> SET TAGS (...) and ALTER COLUMN ... SET TAGS (...).
Snowflake: use ALTER TABLE <table> SET TAG <name> = '<value>'; adjust the macro accordingly.
BigQuery: consider labels or policy tags via separate APIs; the post-hook may call an external procedure instead of SQL.

Example: end-to-end

Add meta in YAML for your model and columns.
Ensure the macro is available and included in the models’ post-hooks.
Run your model; dbt will execute the post-hook and apply tags to the relation and columns.

Reference implementation

set_gdpr_compliance_tag macro (full)

-- Generic GDPR tagging macro for dbt
-- Notes:
-- - Example SQL below uses Databricks/Unity Catalog syntax for tags.
-- - If your warehouse uses a different syntax (e.g., Snowflake), adjust the ALTER statements accordingly.

{% macro set_gdpr_compliance_tag() %}
    {% if execute %}
        {# Table-level meta flag to opt-in the model for GDPR tagging #}
        {% set is_gdpr_model = model.meta.get('is_gdpr_model') %}

        {% if is_gdpr_model %}
            {# Table-level tag #}
            {% set table_tag_name = 'gdpr_deletion' %}
            {% set table_tag_value = 'enabled' %}

            {% set table_level_tag_sql %}
                ALTER TABLE {{ this }} SET TAGS ('{{ table_tag_name }}' = '{{ table_tag_value }}');
            {% endset %}
            {% do run_query(table_level_tag_sql) %}

            {# Column-level tags: set gdpr_column_type when provided #}
            {% set column_tag_name = 'gdpr_column_type' %}
            {% for column in model.columns %}
                {% set col_meta = model.columns.get(column).meta or {} %}

                {# Support either a boolean flag or explicit type string #}
                {% set explicit_type = col_meta.get('gdpr_column_type') %}
                {% set is_flagged = col_meta.get('is_gdpr_column') %}

                {% if explicit_type is string and explicit_type|length > 0 %}
                    {% set column_level_tag_sql %}
                        ALTER TABLE {{ this }}
                        ALTER COLUMN {{ column }} SET TAGS ('{{ column_tag_name }}' = '{{ explicit_type }}');
                    {% endset %}
                    {% do run_query(column_level_tag_sql) %}
                {% elif is_flagged %}
                    {% set column_level_tag_sql_default %}
                        ALTER TABLE {{ this }}
                        ALTER COLUMN {{ column }} SET TAGS ('{{ column_tag_name }}' = 'user_identifier');
                    {% endset %}
                    {% do run_query(column_level_tag_sql_default) %}
                {% else %}
                    {% set column_level_untag_sql %}
                        ALTER TABLE {{ this }}
                        ALTER COLUMN {{ column }} UNSET TAGS ('{{ column_tag_name }}');
                    {% endset %}
                    {% do run_query(column_level_untag_sql) %}
                {% endif %}
            {% endfor %}
        {% endif %}
    {% endif %}
{% endmacro %}

YAML meta example for GDPR

version: 2

models:
  - name: user_events_monthly
    description: Monthly user engagement events used for activity analytics.
    meta:
      is_gdpr_model: true
      retention: 400d
    columns:
      - name: user_id
        description: Anonymized user identifier
        meta:
          gdpr_column_type: user_identifier
      - name: session_id
        description: Analytics session identifier
        meta:
          is_gdpr_column: true
      - name: event_date
        description: Logical partition/event date
      - name: event_type
        description: Type of event

  - name: advertising_campaigns
    description: Campaigns table retained permanently for financial audit.
    meta:

Operational tips

Run this macro via post-hooks to keep tags aligned with each run, especially after schema changes.
Use explicit types like user_identifier, pii_email, or business_identifier to categorize sensitivity.
Keep the tagging macro idempotent and safe for reruns; the provided implementation sets or unsets tags based on current meta.