Warehouse-native

A warehouse-native tool reads and processes data where it already sits, in your own warehouse such as Snowflake, BigQuery or Redshift. A conventional tool asks you to send a copy of your events and customer records to its servers. A warehouse-native one queries your copy instead, so the warehouse stays the single source of truth. For example, a marketing team can build audiences from the same tables the finance team reports from, with no second copy to keep in step. RudderStack is a tool in the library that lists this feature.

Duplicate copies of customer data create mismatched numbers, extra security reviews and extra storage bills. Keeping the data in one place makes definitions of terms such as an active customer consistent, and it keeps data under your own access rules. You need this when you already run a warehouse, when you have strict data residency or privacy requirements, or when several teams argue about whose numbers are right. You can skip it for a small business with no warehouse, where a standard tool with its own storage is simpler.

• Which warehouses are supported, and whether support is complete or limited to certain features.
• Whether any data still leaves your warehouse, and where processing takes place.
• How the cost of queries run in your warehouse is shown and controlled.
• How much SQL or data engineering knowledge your team needs for setup and day-to-day use.