Announcing Snowflake Support: The Data Warehouse Built for the Cloud

Announcing Snowflake Support: The Data Warehouse Built for the Cloud

Snowflake changed the game for data warehousing. By separating compute from storage, they made scalable analytics accessible to everyone—from solo founders to massive teams.

But for many engineers, getting data into Snowflake is still a headache.

You usually have two bad options:

  1. The “Scripting Trap”: You write Python scripts to dump CSVs to an S3 bucket, manage file formats, and write fragile COPY INTO commands. If a column name changes, your script breaks.
  2. The “Usage-Based” Trap: You use a big-box ETL tool that sees “Snowflake” and assumes you have an infinite budget, charging you per row for data you already own.

We built Saddle Data to break this cycle. Today, we’re proud to announce Snowflake is now a supported Destination.

How We Built It (The Engineering View)

We didn’t just wrap a generic JDBC driver. We built this connector specifically to leverage Snowflake’s architecture for performance and efficiency.

1. Simplified Loading

Under the hood, Saddle Data handles the intermediate steps. We manage the batching, file formatting, and loading execution. You don’t need to configure and maintain your own external S3 stage or Azure Blob container just to move a table; we handle the complexity for you.

2. Schema Drift Detection

This is a critical feature for maintaining reliable pipelines. If you add a discount_code column to your production Postgres database, Saddle Data detects the change.

  • Old Way: The pipeline fails. You manually log into Snowflake, run ALTER TABLE, and restart the job.
  • Saddle Way: Depending on your configured policy, we can either automatically migrate the schema (issuing the ALTER TABLE command in Snowflake) or pause the pipeline and notify you to approve the change.

3. Incremental Sync (Using MERGE)

Full refreshes are expensive—both in Saddle Data credits and Snowflake compute credits. Our connector supports our “Incremental Deduped” mode. We track the cursor (like updated_at) and use Snowflake’s efficient MERGE command to update only the records that have changed. This keeps your Snowflake compute costs low and your data fresh.

Why This Matters

We believe you shouldn’t be “taxed” for choosing a powerful data warehouse.

In the current market, moving to Snowflake often signals to vendors that it’s time to increase your price. At Saddle Data, our Snowflake connector is available to our users without hidden “enterprise-only” tax. You get the power of the world’s best data warehouse with the predictability of our standard plans.

Getting Started

The Snowflake connector is live in the app today.

  1. Create a new Destination in the Saddle Data builder.
  2. Select Snowflake.
  3. Enter your Account URL, User, and Warehouse details.
  4. Start syncing.

Ready to build? Log in to your dashboard and give it a spin.