Why Do Snowflake Schema, Table, and Column Names Use Different Casing?

Question

Why do some connections create uppercase schema, table, and column names in Snowflake while others create lowercase names, even when the names are the same in Fivetran?

Environment

Destination: Snowflake

Answer

This depends on the Destination names option you select while creating the connection:

• Fivetran naming: We normalize the case to uppercase. For example, google_sheets becomes GOOGLE_SHEETS.
• Source naming: We preserve the case as it appears in the source. For example, google_sheets remains google_sheets.

The exception is reserved keywords used as identifiers — with Fivetran naming, we convert these to lowercase and quote them instead. For more information, see Reserved keywords.

Therefore, two connections can display the same name in Fivetran but create schemas, tables, or columns with different casing in the Snowflake destination. For more information, see our Naming conventions documentation.

To check which naming option a connection uses:

1. In Fivetran, go to the relevant connection page.
2. Select the Settings tab.
3. Click Edit Connection.
4. Check the Destination names setting.

You can only change the Destination names setting before the connection's initial sync begins. After the initial sync starts, you must create a new connection to use a different setting.