What is a dbt macro?

A macro is a reusable Jinja function defined in macros/*.sql that generates SQL (or snippets) for models, tests, and hooks.

Example

-- macros/cents_to_dollars.sql
{% macro cents_to_dollars(column_name) %}
    ({{ column_name }} / 100.0)
{% endmacro %}

Use it:

select
    order_id,
    {{ cents_to_dollars('amount_cents') }} as amount
from {{ ref('stg_orders') }}

When macros help

• Repeated expressions (surrogate keys, standardized timestamps)
• Adapter-specific SQL differences (dateadd vs date_add)
• Custom generic tests
• DRY patterns shared across many models

When macros hurt

• Hiding simple SQL behind opaque layers beginners cannot read
• Business logic that should be an intermediate model instead
• Over-abstracted frameworks nobody can debug

Mental model

macro  =  function that returns SQL text at compile time
model  =  dataset definition executed in the warehouse

Interview tip: "I extract a macro after the third copy-paste, and I keep business grain as models, not as nested macros."