Legacy snapshot configuration legacy

From dbt versions 1.8 and earlier, you were able to configure snapshots using jinja blocks in your .sql files. This is considered legacy syntax and was replaced with a YAML-based configuration in dbt Cloud's "Latest" release track and dbt v1.9 for faster and more efficient management.

However, there are situations where you might still need to use the legacy syntax for snapshots in any dbt version or release track. This page details how you can use the legacy SQL-based configurations and provides a path to migrate to the more efficient YAML configuration.

For new snapshots, we recommend using these latest YAML-based configs. If applying them to existing snapshots, you'll need to migrate over.

The following outlines the differences between the legacy SQL-based syntax and the updated YAML-based syntax:

 SQL-based snapshot syntax

Legacy syntax for defining snapshots in .sql files within a snapshot Jinja block, typically located in your snapshots directory. Available in dbt v1.8 and earlier.

Use cases

• Defining multiple resources in a single file, though it can significantly slow down parsing and compilation.
• Useful for existing snapshots already using this syntax.
• Suitable for performing very light transformations (but creating a separate ephemeral model for transformations is recommended for better maintainability).

Example

In this example, we created a snapshot in the snapshots directory and used the {{ config() }} block within the jinja block to define the snapshot configuration.

snapshots/orders_snapshot.sql

{% snapshot orders_snapshot %}

{{ config(
    target_database='analytics',
    target_schema='snapshots',
    unique_key='id',
    strategy='timestamp',
    updated_at='updated_at'
) }}

select * from {{ source('jaffle_shop', 'orders') }}

{% endsnapshot %}

 YAML-based snapshot syntax

Updated syntax for defining snapshot configurations in YAML files. Found in snapshots.yml. Available in dbt Cloud's "Latest" release track and dbt v1.9 and later.

Use cases

• More performant and easier to manage.
• Ideal for new snapshots or existing snapshots that need to be migrated.
• Create transformations separate from the snapshot file by creating an ephemeral model and referencing it in the snapshot using the relation field.

Example

In this example, we created a snapshot in the snapshots directory (and separately an ephemeral model in the models directory). We then used the ref function to reference the ephemeral model in the snapshots.yml file in the relation field.

snapshots.yml

snapshots:
 - name: orders_snapshot
   relation: ref('orders_ephemeral')
   config:
     unique_key: id
     strategy: timestamp
     updated_at: updated_at

In this second example, we used the relation field to reference the source table using the source function.

snapshot/orders_snapshot.yaml

snapshots:
  - name: orders_snapshot
    relation: source('jaffle_shop', 'orders')
    config:
      schema: snapshots
      database: analytics
      unique_key: id
      strategy: timestamp
      updated_at: updated_at
      dbt_valid_to_current: "to_date('9999-12-31')"

Snapshot configurations

Although you can use the more performant YAML-based configuration, you might still want to use the legacy configuration to define your snapshots if it suits your needs.

Snapshots can be configured in two main ways:

• Using snapshot-specific configurations
• Or using general configurations

These configurations allow you to control how dbt detects changes in your data and where snapshots are stored. Both types of configurations can coexist in your project in the same config block (or from your dbt_project.yml file or properties.yaml file). You can also configure snapshots using strategies, which define how dbt knows if a row has changed.

Snapshot specific configurations

Snapshot-specific configurations are applicable to only one dbt resource type rather than multiple resource types. You can define these settings within the resource’s file using the {{ config() }} macro (as well as in the project file (dbt_project.yml) or a property file (models/properties.yml for models, similarly for other resources)).

snapshots/orders_snapshot.sql

{ % snapshot orders_snapshot %}

{{ config(
    target_schema="<string>",
    target_database="<string>",
    unique_key="<column_name_or_expression>",
    strategy="timestamp" | "check",
    updated_at="<column_name>",
    check_cols=["<column_name>"] | "all"
    invalidate_hard_deletes : true | false
) 
}}

select * from {{ source('jaffle_shop', 'orders') }}

{% endsnapshot %}

General configuration

Use general configurations for broader operational settings applicable across multiple resource types. Like resource-specific configurations, these can also be set in the project file, property files, or within resource-specific files using a config block.

snapshots/snapshot.sql

{{ config(
    enabled=true | false,
    tags="<string>" | ["<string>"],
    alias="<string>", 
    pre_hook="<sql-statement>" | ["<sql-statement>"],
    post_hook="<sql-statement>" | ["<sql-statement>"]
    persist_docs={<dict>}
    grants={<dict>}
) }}

Snapshot strategies

Snapshot "strategies" define how dbt knows if a row has changed. There are two strategies built-in to dbt that require the strategy parameter:

• Timestamp — Uses an updated_at column to determine if a row has changed.
• Check — Compares a list of columns between their current and historical values to determine if a row has changed. Uses the check_cols parameter.

Timestamp

The timestamp strategy uses an updated_at field to determine if a row has changed. If the configured updated_at column for a row is more recent than the last time the snapshot ran, then dbt will invalidate the old record and record the new one. If the timestamps are unchanged, then dbt will not take any action.

Example

snapshots/timestamp_example.sql

{% snapshot orders_snapshot_timestamp %}

    {{
        config(
          target_schema='snapshots',
          strategy='timestamp',
          unique_key='id',
          updated_at='updated_at',
        )
    }}

    select * from {{ source('jaffle_shop', 'orders') }}

{% endsnapshot %}

Check

The check strategy is useful for tables which do not have a reliable updated_at column. It requires the check_cols parameter, which is a list of columns within the results of your snapshot query to check for changes. Alternatively, use all columns using the all value (however this may be less performant).

Example

snapshots/check_example.sql

{% snapshot orders_snapshot_check %}

    {{
        config(
          strategy='check',
          unique_key='id',
          check_cols=['status', 'is_cancelled'],
        )
    }}

    select * from {{ source('jaffle_shop', 'orders') }}

{% endsnapshot %}

Examples

 Check a list of columns for changes

snapshots/check_example.sql

{% snapshot orders_snapshot_check %}

    {{
        config(
          strategy='check',
          unique_key='id',
          check_cols=['status', 'is_cancelled'],
        )
    }}

    select * from {{ source('jaffle_shop', 'orders') }}

{% endsnapshot %}

 Check all columns for changes

snapshots/check_example.sql

{% snapshot orders_snapshot_check %}

    {{
        config(
          strategy='check',
          unique_key='id',
          check_cols='all',
        )
    }}

    select * from {{ source('jaffle_shop', 'orders') }}

{% endsnapshot %}

Configure snapshots

In dbt versions 1.8 and earlier, snapshots are select statements, defined within a snapshot block in a .sql file (typically in your snapshots directory or any other directory). You'll also need to configure your snapshot to tell dbt how to detect record changes.

The following table outlines the configurations available for snapshots in versions 1.8 and earlier:

Config	Description	Required?	Example
target_database	The database that dbt should render the snapshot table into	No	analytics
target_schema	The schema that dbt should render the snapshot table into	Yes	snapshots
strategy	The snapshot strategy to use. One of timestamp or check	Yes	timestamp
unique_key	A primary key column or expression for the record	Yes	id
check_cols	If using the check strategy, then the columns to check	Only if using the check strategy	["status"]
updated_at	If using the timestamp strategy, the timestamp column to compare	Only if using the timestamp strategy	updated_at
invalidate_hard_deletes	Find hard deleted records in source, and set dbt_valid_to current time if no longer exists	No	True

• A number of other configurations are also supported (like, tags and post-hook), check out the full list here.
• Snapshots can be configured from both your dbt_project.yml file and a config block, check out the configuration docs for more information.
• Note: BigQuery users can use target_project and target_dataset as aliases for target_database and target_schema, respectively.

Add snapshot to a project

To add a snapshot to your project:

1. Create a file in your snapshots directory with a .sql file extension. For example,snapshots/orders.sql
2. Use a snapshot block to define the start and end of a snapshot:

snapshots/orders_snapshot.sql

{% snapshot orders_snapshot %}

{% endsnapshot %}

3. Write a select statement within the snapshot block (tips for writing a good snapshot query are below). This select statement defines the results that you want to snapshot over time. You can use sources or refs here.

snapshots/orders_snapshot.sql

{% snapshot orders_snapshot %}

select * from {{ source('jaffle_shop', 'orders') }}

{% endsnapshot %}

4. Check whether the result set of your query includes a reliable timestamp column that indicates when a record was last updated. For our example, the updated_at column reliably indicates record changes, so we can use the timestamp strategy. If your query result set does not have a reliable timestamp, you'll need to instead use the check strategy — more details on this in the next step.

5. Add configurations to your snapshot using a config block. You can also configure your snapshot from your dbt_project.yml file (docs).

snapshots/orders_snapshot.sql

{% snapshot orders_snapshot %}

{{
    config(
      target_database='analytics',
      target_schema='snapshots',
      unique_key='id',

      strategy='timestamp',
      updated_at='updated_at',
    )
}}

select * from {{ source('jaffle_shop', 'orders') }}

{% endsnapshot %}

6. Run the dbt snapshot command. For our example, a new table will be created at analytics.snapshots.orders_snapshot. You can change the target_database configuration, the target_schema configuration and the name of the snapshot (as defined in {% snapshot .. %}) will change how dbt names this table.

dbt snapshot
Running with dbt=1.8.0

15:07:36 | Concurrency: 8 threads (target='dev')
15:07:36 |
15:07:36 | 1 of 1 START snapshot snapshots.orders_snapshot...... [RUN]
15:07:36 | 1 of 1 OK snapshot snapshots.orders_snapshot..........[SELECT 3 in 1.82s]
15:07:36 |
15:07:36 | Finished running 1 snapshots in 0.68s.

Completed successfully

Done. PASS=2 ERROR=0 SKIP=0 TOTAL=1

7. Inspect the results by selecting from the table dbt created. After the first run, you should see the results of your query, plus the snapshot meta fields as described earlier.

8. Run the dbt snapshot command again, and inspect the results. If any records have been updated, the snapshot should reflect this.

9. Select from the snapshot in downstream models using the ref function.

models/changed_orders.sql

select * from {{ ref('orders_snapshot') }}

10. Snapshots are only useful if you run them frequently — schedule the snapshot command to run regularly.

Examples

This section outlines some examples of how to apply configurations to snapshots using the legacy method.

 Apply configurations to one snapshot only

Use config blocks if you need to apply a configuration to one snapshot only.

snapshots/postgres_app/orders_snapshot.sql

{% snapshot orders_snapshot %}
    {{
        config(
          unique_key='id',
          strategy='timestamp',
          updated_at='updated_at'
        )
    }}
    -- Pro-Tip: Use sources in snapshots!
    select * from {{ source('jaffle_shop', 'orders') }}
{% endsnapshot %}

 Using the updated_at parameter

The updated_at parameter is required if using the timestamp strategy. The updated_at parameter is a column within the results of your snapshot query that represents when the record row was last updated.

snapshots/orders.sql

{{ config(
  strategy="timestamp",
  updated_at="column_name"
) }}

Examples

• Using a column name updated_at:

  snapshots/orders.sql

  {% snapshot orders_snapshot %}
  
  {{
      config(
        target_schema='snapshots',
        unique_key='id',
  
        strategy='timestamp',
        updated_at='updated_at'
      )
  }}
  
  select * from {{ source('jaffle_shop', 'orders') }}
  
  {% endsnapshot %}

• Coalescing two columns to create a reliable updated_at column:

  Consider a data source that only has an updated_at column filled in when a record is updated (so a null value indicates that the record hasn't been updated after it was created).

  Since the updated_at configuration only takes a column name, rather than an expression, you should update your snapshot query to include the coalesced column.

  snapshots/orders.sql

  {% snapshot orders_snapshot %}
  
  {{
      config(
        target_schema='snapshots',
        unique_key='id',
  
        strategy='timestamp',
        updated_at='updated_at_for_snapshot'
      )
  }}
  
  select
      *,
      coalesce(updated_at, created_at) as updated_at_for_snapshot
  
  from {{ source('jaffle_shop', 'orders') }}
  
  {% endsnapshot %}

 Using the unique_key parameter

The unique_key is a column name or expression that is unique for the inputs of a snapshot. dbt uses unique_key to match records between a result set and an existing snapshot, so that changes can be captured correctly.

snapshots/orders.sql

{{ config(
  unique_key="column_name"
) }}

Examples

• Using an id column as a unique key

  snapshots/orders.sql

  {{
      config(
        unique_key="id"
      )
  }}

  You can also write this in YAML. This might be a good idea if multiple snapshots share the same unique_key (though we prefer to apply this configuration in a config block, as above).

• Using a combination of two columns as a unique key

  This configuration accepts a valid column expression. As such, you can concatenate two columns together as a unique key if required. It's a good idea to use a separator (like, '-') to ensure uniqueness.

  snapshots/transaction_items_snapshot.sql

  {% snapshot transaction_items_snapshot %}
  
      {{
          config(
            unique_key="transaction_id||'-'||line_item_id",
            ...
          )
      }}
  
  select
      transaction_id||'-'||line_item_id as id,
      *
  from {{ source('erp', 'transactions') }}
  
  {% endsnapshot %}

  Though, it's probably a better idea to construct this column in your query and use that as the unique_key:

  snapshots/transaction_items_snapshot.sql

  {% snapshot transaction_items_snapshot %}
  
      {{
          config(
            unique_key="id",
            ...
          )
      }}
  
  select
      transaction_id || '-' || line_item_id as id,
      *
  from {{ source('erp', 'transactions') }}
  
  {% endsnapshot %}

Migrate legacy snapshot configs

This section outlines the steps you need to follow to migrate legacy jinja block snapshot configurations into the updated YAML-based configuration format.

Why use the updated YAML spec?

• YAML-based configurations are processed faster by dbt, leading to improved performance, especially during parsing and compilation.
• Centralizing configuration in YAML makes it easier to manage and update snapshot settings without editing the SQL logic directly.
• YAML configuration aligns snapshot definitions with other dbt resources, such as models and seeds, leading to a more consistent project structure.

Considerations

• In versions prior to v1.9, the target_schema (required) and target_database (optional) configurations defined a single schema or database to build a snapshot across users and environment. This created problems when testing or developing a snapshot, as there was no clear separation between development and production environments.
• In v1.9, target_schema became optional, allowing snapshots to be environment-aware.
• By default, without target_schema or target_database defined, snapshots now use the generate_schema_name or generate_database_name macros to determine where to build.
• Developers can still set a custom location with schema and database configs, consistent with other resource types.

How to migrate

The latest YAML-based configuration syntax is best suited for new snapshots. If you're migrating existing snapshots, consider the following steps:

1. Migrate the previous snapshot to the new table schema and values.

   • Create a backup copy of your snapshots.
   • Use alter statements as needed (or a script to apply alter statements) to ensure table consistency.

2. Convert any configurations currently written within the jinja block (like unique_key, strategy, updated_at, and so on) into the YAML file structure, one at a time and testing as you go.

   The configurations are structured similarly to how you would define a model in dbt_project.yml.:

   snapshots.yml

   snapshots:
     - name: orders_snapshot
       schema: snapshots
       unique_key: id
       strategy: timestamp
       updated_at: updated_at

   Note: The unique_key, strategy, and updated_at fields must match the settings previously defined in your jinja block.

3. Before removing the old jinja block, run the dbt snapshot command using the new YAML configuration to confirm that the snapshot behaves as expected.

   • Verify that the data is processed correctly (for example, no data loss or incorrect records).
   • Make sure the performance is either the same or improved compared to the old configuration.
   • After running the new snapshot, inspect the snapshot tables in your data warehouse to confirm the new snapshot records match the old data.

4. Once you’ve confirmed that the new YAML configuration works properly, safely remove the old snapshot jinja block from your .sql file. This keeps your codebase clean and fully migrated to the new method.

5. If your snapshots require more complex transformations, consider using an ephemeral model to handle the transformations before referencing it in the snapshot. An ephemeral model can encapsulate transformations and simplify the snapshot query itself.

   Example of using an ephemeral model:

   models/ephemeral/orders_ephemeral.sql

   {{
     config(materialized='ephemeral')
   }}
   select * from {{ source('jaffle_shop', 'orders') }}

   Example of the snapshot YAML configuration referencing the ephemeral model:

   snapshots.yml

   snapshots:
     - name: orders_snapshot
       relation: ref('orders_ephemeral')
       target_schema: snapshots
       unique_key: id
       strategy: timestamp
       updated_at: updated_at

Full migration example

Here’s a complete example of migrating from a legacy jinja block snapshot to a YAML-based snapshot configuration:

Legacy method (jinja block)

snapshots/orders_snapshot.sql

{% snapshot orders_snapshot %}
{{
    config(
      target_schema='snapshots',
      unique_key='id',
      strategy='timestamp',
      updated_at='updated_at',
    )
}}
select * from {{ source('jaffle_shop', 'orders') }}
{% endsnapshot %}

Recommended method (YAML configuration)

snapshots.yml

snapshots:
  - name: orders_snapshot
    schema: snapshots
    unique_key: id
    strategy: timestamp
    updated_at: updated_at

By following these steps, you can smoothly transition from legacy jinja-based snapshots to the modern, more efficient YAML-based configurations.