Fivetran to ClickHouse

Synchronize data from over 400 sources to Propel’s Serverless ClickHouse.

Request access to Fivetran destination

Contact us if you are interesting in getting early access to the Fivetran destination to sync your data to Propel.

Get started with Fivetran

Step-by-step instructions to connect any Fivetran source to Propel.

Check out the Propel Fivetran destination docs.

​

Features

Data Pools created with the Fivetran destination are Webhook Data Pools that support the following features:

Feature name	Supported	Notes
Syncs inserts, updates, and deletes	✅	Propel processes inserts, updates, and deletes sent from Fivetran.
Re-sync	✅	You can trigger a resync in Fivetran.
Configurable sync interval	✅	You can configure sync interval in Fivetran.
Sync Pausing / Resuming	✅	You can pause / resume syncs in Fivetran.
Schema Changes: Add Column	✅	When Fivetran detects a column was added in your source, it will automatically add that column in Propel.
Schema Changes: Change column types	❌	Changing columns types is not supported as it is a breaking schema change.
Schema Changes: Delete column	❌	Deleting columns is not supported as it is a breaking schema change.
Batch Delete API	✅	See Batch Delete API.
Batch Update API	✅	See Batch Update API.
API configurable	✅	See API docs.
Terraform configurable	✅	See Terraform docs.

​

How does the Fivetran destination work?

When syncing a Fivetran source, Propel creates a Webhook Data Pool for each table. For example, a Google Search Console Fivetran source would result in multiple Data Pools:

Each Data Pool mirrors its source table’s structure, with additional columns:

Column	Type	Description
_propel_received_at	TIMESTAMP	UTC timestamp of Propel’s receipt
_propel_payload	JSON	Event’s JSON payload
_fivetran_synced	TIMESTAMP	UTC timestamp of Fivetran’s sync initiation

Fivetran performs initial and subsequent syncs based on your configured frequency, sending data to Propel’s Webhook Data Pool API for processing and querying via SQL and API.

​

Key configurations

• Default Timestamp: First timestamp column or _propel_received_at if none exists.
• Primary Key: Source table’s primary key or a Fivetran-generated unique ID if absent.
• Sort Key: Combination of primary key and default timestamp.

​

Data handling

• Updates: Identified and applied using the primary key.
• Deletes: Processed based on delete events from Fivetran using the primary key.

​

Schema changes

Propel automatically adds new columns detected by Fivetran. Column type changes or deletions are not supported due to their breaking nature.

​

Data types

The following table illustrates how we transform your Fivetran data types into Propel-supported types:

FIVETRAN DATA TYPE	PROPEL DATA TYPE
BOOLEAN	BOOLEAN
SHORT	INT16
INT	INT32
LONG	INT64
BIGDECIMAL	DOUBLE
FLOAT	FLOAT
DOUBLE	DOUBLE
LOCALDATE	DATE
LOCALDATETIME	TIMESTAMP
INSTANT	TIMESTAMP
STRING	STRING
XML	STRING
JSON	JSON
BINARY	STRING

​

Transforming data

Once your data is in a Webhook Data Pool, you can use Materialized Views to:

• Flatten nested JSON into tabular form
• Flatten JSON array into individual rows
• Combine data from multiple source Data Pools through JOINs
• Calculate new derived columns from existing data
• Perform incremental aggregations
• Sort rows with a different sorting key
• Filter out unnecessary data based on conditions
• De-duplicate rows