Short answer: Use PostgreSQL or MySQL for structured relational data and controlled queries. Use Google Sheets when people need to review or edit a small table directly. Match the node to the data owner and failure tolerance.

What is PostgreSQL best suited to?

PostgreSQL fits relational records, joins, transactions, and applications that already use the database. Define indexes for fields used in workflow filters. Use parameters for changing values instead of building SQL with string concatenation.

Test the query with a limited result set first. Check the returned column names before mapping them into later nodes.

When does MySQL make sense?

MySQL is a practical choice when the existing application already stores its data there. Reuse the database’s naming, permissions, and transaction rules. Do not assume that a query written for one SQL dialect behaves identically in the other.

Limit the workflow account to the tables and operations it needs. Separate read and write credentials when the process requires that boundary.

When is Google Sheets the right tool?

Sheets works well for a small operational list, review queue, or hand-maintained lookup. It gives non-technical users a visible table. It is a poor fit for high-volume concurrent writes or strict relational integrity.

Use a stable key column. Decide how the workflow handles a missing row, duplicate key, and manual edit. Protect columns that the workflow owns.

How should connection tests be run?

Test credentials first. Read one known record. Write a non-critical test value. Confirm the result in the source system. Then test an empty result and a rejected permission.

For storage selection, compare the Data Store and PostgreSQL guide. For the wider connector map, see the integration node catalog.