Firestore Data Modeling Patterns FAQs

Question 1

What is the primary difference between transactions and batched writes?

Accepted Answer

Transactions are required when a write operation depends on the current state of a document (read-modify-write). Batched writes are for multiple independent writes that must succeed or fail together, and they are generally faster and work offline.

Question 2

Does Firestore support cascading deletes for subcollections?

Accepted Answer

No. Deleting a parent document does not automatically delete documents in its subcollections. You must implement your own cleanup logic, typically using a Cloud Function, to prevent orphaned data.

Question 3

When should I use a subcollection instead of a root collection?

Accepted Answer

Use subcollections when data is tightly coupled to a specific parent document and you don't need to query that data globally. Use root collections for many-to-many relationships or when you need to query across all parents.

Question 4

How do I handle many-to-many relationships in Firestore?

Accepted Answer

The best practice is usually to use root collections and store arrays of document IDs (references) or use a junction collection, depending on the scale of the data and query requirements.

Firestore Data Modeling Patterns

Key Features

Use Cases

Firestore Data Modeling Patterns

Key Features

Use Cases