Pre-Built Data Models
Fivetran offers pre-built, dbt Core*-compatible data models for our most popular connectors. Our pre-built models produce clear, comprehensible tables that power your reports, visualizations, and dashboards.
You can either add the models in the Fivetran dashboard (Quickstart data models) or import them into your dbt project (Fivetran data models).
Supported destinations
The following destinations support Fivetran data models:
1Databricks on Azure destinations with OAuth authentication don't support Quickstart data models.
Supported destinations vary by data model. Check the prerequisites section of your chosen data model's documentation to find the exact list of destinations it supports. You do not need to enable Transformations for dbt Core* to use our models, but we recommend it.
If you'd like more destinations to support Fivetran data models, contact Support to submit your request.
How it works
We built data models that transform your data into analytics-ready tables. The models transform Fivetran’s normalized schemas into analytics-ready datasets for each connector's most common use cases.
To use our pre-built data models, you can set them up directly in the Fivetran dashboard (Quickstart data models) or add them to your dbt project (Fivetan data models).
Quickstart data models
Quickstart data models allows you to use our pre-built data models without building out your own dbt project. You manage Quickstart data models directly in the Fivetran dashboard.
Once you set up a Quickstart data model, you can select the output models. We then run the transformations according to your selected schedule. Note that each Quickstart transformation only includes the selected output models and the intermediate data models that are required for the selected output models. You can view, edit, and delete each transformation from the Transformations page in the dashboard.
We charge for all model runs, including both output model runs and intermediate model runs required to generate the output models. See our How Many Models Will Run for My Quickstart Transformation? FAQ to learn how to check the exact number of data models we run for your selected output models.
Limitations
- For single-source models, you can only associate one Quickstart data model with one connection. However, you can bulk-create Quickstart data models — one for each connection.
- You can only have one multi-source model of a given type per destination.
- You must use the same destination credentials for your Quickstart data model as you use for your connector. For example, if your connector loads data into the Test destination, you must use the Test destination to transform the data.
Fivetran data models
After Fivetran syncs your data to your destination, our data models transform your data and create analytics-ready tables in your destination.
Some data models have all necessary files combined into a single model. However, most Fivetran data models consist of two smaller models:
- The source models expose, document and standardize the underlying Fivetran schemas being created in the destination.
- The transform models produce analytics-ready data models for each source.
Source models
Source models add column-level tests and perform minor transformations on your destination data (for example, renaming columns). They are designed to help you test and clean your data and to create a dictionary of your data. They also prepare your data for our transform models, if you choose to use them.
Each source model does the following:
- Adds descriptions to the tables and columns that Fivetran has synced.
- Adds column-level testing where applicable. For example, we test all primary keys for uniqueness and non-null values.
- Models the staging tables that our transform models require.
Transform models
Transform models transform your data into analytics-ready models. They are designed to help you understand and analyze your data. They're especially helpful if you're using your connector for the first time or haven't spent much time modeling your data.
Each transform model does the following:
- Automatically installs and runs the corresponding source model.
- Performs a series of SQL transformations on the source model's staging tables. The specific transformations vary by connector.
- Produces models that address common analytics scenarios for your connector.
Available data models
We provide both single-source and multi-source data models. Multi-source data models model and merge data from connections of similar type.
Fivetran has data models for the following connectors:
Some data models have all necessary files combined into a single model. However, most Fivetran data models consist of two smaller models - source and transform.
If you'd like Fivetran to create a data model for a connector that's not listed here, contact Support to submit your request.
Multi-source models
We support the following multi-source data models:
- Ad Reporting: Aggregates data from our most popular advertising connectors.
- App reporting: Aggregates data from our most popular app platform connectors.
- Social Media Reporting: Aggregates data from our most popular social media connectors.
- Unified RAG: Generates an end model containing unstructured document data to be used for Retrieval Augmented Generation (RAG) applications leveraging Large Language Models (LLMs).
- Shopify Holistic: Aggregates Shopify data with data from other marketing connectors. The Shopify Holistic model supports combining Shopify data with email and SMS marketing data from Fivetran's Klaviyo data model.
Schema naming
Data model schema names vary depending on the connector type. See the Change the Build Schema or Changing the Build Schema section of the relevant Transformation data model's documentation.
How to use Quickstart data models
You must have an active connection of the relevant type to use a Quickstart data model. For example, to use the HubSpot Quickstart data model, you must have an active HubSpot connection.
To learn how to use Quickstart data models in the Fivetran dashboard, see our Quickstart management guide.
How to use Fivetran data models
You must have an active connection of the relevant type to use a Fivetran data model. For example, to use the HubSpot data model, you must have an active HubSpot connection.
To add Fivetran data models to your dbt project, do the following:
- Go to Fivetran's dbt hub.
- Click on the name of your connector.
- Follow the instructions in the Installation section at the top of the page.
- Run the
dbt depscommand. The data models for your connector will automatically install in thedbt_modulesfolder. - If you're ready to start your data models, run the
dbt runcommand.
* dbt Core is a trademark of dbt Labs, Inc. All rights therein are reserved to dbt Labs, Inc. Fivetran Transformations is not a product or service of or endorsed by dbt Labs, Inc.