Skip to main content
Connect Firestore to let Datost query your NoSQL document collections in plain English. Ideal for product teams whose operational data lives in Firestore instead of a SQL warehouse.
Datost reads Firestore live on every question using a SQL-like query layer, it does not copy or cache documents. Results stream back into Slack or the web app as soon as the query returns.

​
What you get

  • Ask questions like @Datost how many users signed up this week? and have them answered against live Firestore collections.
  • Automatic schema inference by sampling documents, including nested maps, arrays, timestamps, GeoPoints, and DocumentReferences.
  • Support for root collections, subcollection paths (users/abc123/orders), and collection group queries across every subcollection with the same name.

​
Prerequisites

Before connecting, make sure you have:
  • A Google Cloud project with Firestore in Native Mode enabled. Datastore Mode is not supported.
  • The Cloud Datastore Viewer (roles/datastore.viewer) role on the project, or a custom role with datastore.documents.get and datastore.documents.list.
  • For service-account auth: permission to create a service account key in the project.
If your project uses Firestore in Datastore Mode, the connection test will fail with a clear error. You’ll need to create a new project with Native Mode, or connect a different one.

​
Authentication methods

Fastest path. Datost requests the datastore and cloudplatformprojects.readonly scopes and uses the signed-in user’s permissions. Refresh tokens are encrypted per-organization at rest.
Recommended for shared org-wide access. Create a service account in GCP, grant it roles/datastore.viewer, generate a JSON key, and paste it when connecting.

​
Connect Firestore

1

Open the admin panel

In the Datost web app, go to Data Sources and click Add data source.
2

Pick Firestore

Select Firestore from the list of warehouse types.
3

Choose an auth method

Toggle between Sign in with Google and Service Account at the top of the form.
4

Authenticate

  • OAuth: click Sign in with Google and approve access. Datost filters out Datastore Mode projects automatically, then lets you pick a GCP Project ID from the dropdown.
  • Service account: paste the JSON key into the Service Account Key field and enter the Project ID.
5

Test and save

Click Test Connection. Datost lists the root collections it can see as a sanity check. If it passes, name the connection and save.

​
How querying works

Datost translates questions into a SQL-like syntax that compiles to native Firestore queries:
  • SELECT * FROM users WHERE plan = 'pro' ORDER BY createdAt DESC LIMIT 100
  • SELECT * FROM users/abc123/orders WHERE status IN ('paid', 'shipped')
  • SELECT * FROM COLLECTION_GROUP(orders) WHERE total > 500
Supported operators include ==, !=, <, <=, >, >=, IN, NOT IN, ARRAY_CONTAINS, and ARRAY_CONTAINS_ANY. Nested fields via dot notation (properties.status) work in both WHERE and ORDER BY.

​
Limits vs SQL warehouses

Firestore is a document database, not a relational one. Datost cannot perform joins, aggregations (GROUP BY, SUM, AVG), or subqueries against Firestore. Ask questions that map to single-collection reads.
  • No joins across collections. Use collection groups or denormalize at write time.
  • Rows are capped at 1,000 per query; schemas are inferred from a sample of up to 100 documents per collection so rare fields may be missed.
  • Firestore composite-index requirements still apply: queries combining multiple range filters or non-indexed ORDER BY fields will fail until you add the matching index in GCP.