Google Cloud Realtime Database n8n integrationAutomate Google Cloud Realtime Database with n8n.
Where should a workflow park its JSON so the next step, or your app, can read it back? The Google Cloud Realtime Database n8n integration puts 5 operations on a single node: write, append, update, read, delete. There is no trigger node, so the run always starts somewhere else.
Verified Trustpilot reviews · AI, automation & growth agency
What does the Google Cloud Realtime Database n8n integration actually do?
Google Cloud Realtime Database is the Firebase store that keeps one JSON tree in sync between a backend and the apps that read it. The n8n node talks to that tree through the Realtime Database API, with 5 operations on a single resource: write a record, append to a list, update fields, read a path, delete a path. You choose the Firebase project, you name the attributes to save, and the workflow handles the round trip.
Start with the case the official catalog already shows: a workflow that collects and stores. Template 787 tracks the position of the ISS, where an HTTP Request node fetches the coordinates and the Google Cloud Realtime Database node appends them with default.push. Every run adds an entry, and the tree becomes a log your app can read without a second API in the middle.
The second case runs the other way. Your app already reads the tree, so anything a workflow writes shows up on the client side on its next read. A support status, a feature flag, a counter for a dashboard: drop it at a known location with default.create, refresh it later with default.update, and no one has to ship a new backend route for it.
Third case, the unglamorous one that saves the most time: housekeeping. default.get pulls a branch of the tree into the workflow, a filter step keeps what matters, the rows land in Google Sheets for the weekly report, and default.delete clears what has already been archived.
Now the honest part. These 5 operations cover reading and writing, and nothing else: create, delete, get, push, update. default.get returns a path, or the whole database object when the path is left blank, and whatever the workflow does with that result happens in the nodes that follow. When you need an endpoint of the Realtime Database API that the node does not expose, the HTTP Request node calls it directly and reuses the same Google credential through its predefined authentication. And when the job really calls for queries or joins, a relational store like Postgres or Supabase sits next to it happily, since a workflow can write to both.
The other limit sits at the start of the workflow. This tool has no trigger node in n8n, so nothing here watches the tree for changes. A run begins with a Schedule Trigger at a fixed interval, with an n8n Webhook, which is a URL your app calls, or with the trigger of another tool such as Airtable. Worth saying out loud: this is not Google Cloud Firestore, which is a different node with a different data model. If the n8n editor itself is still new to you, the n8n training covers the parts this page assumes.
How do you connect the node to your Firebase project?
- 01
Create the Google credential in n8n
A credential is the set of connection details n8n stores once and reuses in every workflow. On n8n Cloud, the fastest route is Managed OAuth2: open the credential screen, click Sign in with Google, and nothing else is needed in the Google Cloud Console. Self-hosted instances do not get Managed OAuth2, so use the dropdown to switch to Custom OAuth2, which asks for a Client ID and a Client Secret you create in the next step.
- 02
Set up the OAuth client in Google Cloud Console
Create a Google Cloud project, enable the APIs you need, then open APIs & Services > OAuth consent screen and fill in the App name and the User support email. For Audience, pick Internal for your own workspace or External for any Google account. Under Branding, add an authorized domain:
n8n.cloudon Cloud, the domain of your instance when self-hosting. - 03
Finish the client and point the node at a project
In Credentials, select + Create credentials > OAuth client ID, choose Web application as the application type, then copy the OAuth Redirect URL shown in your n8n credential into Authorized redirect URIs. Service accounts are documented too, with a Service Account Email and a Private Key taken from the downloaded JSON file, quotes excluded. Back in the node, the Project Name or ID dropdown then lists your projects.
The 5 operations of the node, one by one
The Google Cloud Realtime Database node exposes 5 operations. For each one: the node as you configure it in n8n, the required fields, and our field notes.
Write data to a database
createWhat you see in n8n
Notes & use cases
Saves the incoming item into the database of the Firebase project you select, so the record is there for whatever reads next.
Key parameters
- Project Name or ID: the project as displayed in the firebase console URL. Pick it from the list, or set it with an expression such as
{{ $json.projectId }}when the project comes from an earlier node. - Columns / Attributes: the comma separated fields to save, in the shape
age, name, city. Only what you name here is written, so a field missing from that list never lands in the tree.
Delete data from a database
deleteWhat you see in n8n
Notes & use cases
Removes the data held in the database of the selected project. Nothing comes back from the node to undo it, so this is the operation to test on a throwaway project first.
Key parameters
- Project Name or ID: the only required field here, taken from the firebase console URL. An expression works as well, which makes it easy to point a cleanup workflow at a staging project rather than the live one.
Get a record from a database
getWhat you see in n8n
Notes & use cases
Reads back what is stored and hands it to the workflow as JSON, ready for a filter, a template or another node.
Key parameters
- Project Name or ID: the Firebase project to read from, chosen in the list or built with an expression.
- Object Path: the location to read, written without a trailing
.json, for example/app/users. Left blank, the node returns the whole database object, which gets heavy fast on a real tree.
Append to a list of data
pushWhat you see in n8n
Notes & use cases
Adds the incoming item to a list instead of replacing what is already there, which is what makes repeated runs stack up rather than overwrite each other.
Key parameters
- Project Name or ID: the target project, as displayed in the firebase console URL.
- Columns / Attributes: the fields carried over from the incoming item, listed like
age, name, city, one entry per execution.
Update item in a database
updateWhat you see in n8n
Notes & use cases
Changes the attributes you name on data that already exists, rather than writing a fresh record over the top of it.
Key parameters
- Project Name or ID: the Firebase project holding the item, picked from the list or resolved by expression.
- Columns / Attributes: the comma separated fields to refresh, such as
age, name, city. Name only the ones that changed, since this list is what the node sends.
Need help automating Google Cloud Realtime Database with n8n?
A person reads every message.
Questions people ask next
01Is the Google Cloud Realtime Database n8n integration free?
02What credentials does the node need?
03What are the limits of the Google Cloud Realtime Database node?
04Does the integration react in real time to a change in the database?
05n8n or Make for this database?
Get our weekly integration tips.
No spam. Unsubscribe anytime.
