Dune → Snowflake via dbt

Blockchain analytics starts with raw records such as transfers, trades, and smart-contract calls; Dune’s new setup lets teams model that data with dbt and push the result into Snowflake. (dune.com) Dune said in a blog post published last week that a dbt project can run directly against Dune’s onchain datasets, then use a dbt post-hook to register the output as a synced table for Snowflake or Google BigQuery. The company said Dune handles the incremental syncs after that. (dune.com) Dune’s documentation now lists “dbt Connector,” “dbt to Datashare,” and Snowflake under its data engineering tools, alongside a Datashare product that exports datasets into customer warehouses. Dune’s homepage says the platform covers more than 100 blockchains, while its Snowflake Marketplace listings show packaged datasets for chains including Ethereum, Base, Polygon, Scroll, and cross-chain activity. (docs.dune.com) (dune.com) (app.snowflake.com) dbt, short for data build tool, is the layer many analytics teams use to turn raw tables into cleaned, tested models. dbt Labs’ Snowflake guides describe the standard pattern: connect dbt to Snowflake, run SQL models, and materialize the results into warehouse tables or views for downstream reporting. (docs.getdbt.com 1) (docs.getdbt.com 2) The change here is where the heavy lifting happens. Instead of first copying large amounts of blockchain data into Snowflake and only then transforming it, Dune is pitching a flow where engineers filter, aggregate, and test the data on Dune first, then ship the smaller finished tables into the warehouse. (dune.com) Dune framed that as a cost and workflow issue, especially for data-heavy chains such as Solana, where moving raw data can be expensive. The company said the synced tables arrive in Snowflake “ready to join” with offchain sources such as customer records, payments data, and product analytics. (dune.com) Snowflake already distributes Dune datasets through its Marketplace, which means the new pattern extends an existing relationship rather than creating a new one. Marketplace pages for Dune products describe chain-specific and cross-chain datasets that customers can try inside Snowflake. (app.snowflake.com 1) (app.snowflake.com 2) Dune introduced the dbt Connector at DuneCon 2025, where it described a broader push from dashboarding into full data engineering workflows. In that pitch, Dune said crypto teams no longer need to maintain their own ingestion and storage systems for blockchain data before doing warehouse-style modeling. (dune.com) The social post that circulated this pattern captured a familiar analytics-engineering idea in crypto form: transform data near the source, then move the curated tables into the warehouse where the rest of the business already works. (dune.com)