Connect Google BigQuery

The Google Cloud roles you need, the shared and per-user sign-in choices, whose account the connection queries as, how live queries are billed, and how to fix sign-in and permission errors.

6 min readUpdated October 30, 2026

Connect BigQuery with your Google account and Golden queries it live. Nothing is copied out: every chart runs as a BigQuery job in your project. This page covers the Google Cloud setup, what happens after you connect, and how to read the errors.

Before you start

The person who signs in needs two things in Google Cloud:

BigQuery Job User on the project you enter as Project ID. Golden runs every query as a job in that project, so that's also where the cost lands.
BigQuery Data Viewer on the datasets you want to analyze.

Golden asks Google for read-only BigQuery access, nothing else. It can't change your data or touch other Google services.

When you create a new BigQuery connection, the dialog offers two ways to sign in:

Google sign-in (shared by everyone) - the connection queries as the one person who signed in, so everyone with access to the connection reads data through that account. This is the default.
Google sign-in (per-user) - each member signs in with their own Google account, and their queries run as them. The customer's own BigQuery permissions then decide what each person sees.

Heads up: A shared BigQuery connection queries as whoever signed in. Save it to a shared space and everyone in that space reads data through that one account, so sign in with an account whose access you're happy to share. Pick per-user instead when each member should query with their own Google access.

Note: The per-user choice is offered only while creating a new connection. Editing an existing connection, and the Create Analysis flow's quick connect, use the shared sign-in.

Connect

In a space, click Create and choose Saved connection, then pick Google BigQuery.
Enter the Project ID (lowercase, for example my-analytics-123).
Choose a sign-in method - Google sign-in (shared by everyone) or Google sign-in (per-user).
Click Connect with Google and approve the read-only access. Google shows this screen every time you connect or reconnect. With per-user sign-in, each teammate signs in the first time they open the connection.

More detail on the dialog is in Set up a warehouse connection.

Pick your data

After you save, the import wizard opens with the title "Import Schema from <connection>". It has three tabs: Browse Tables, Query Builder, and Custom SQL.

Browse Tables: search with "Search tables, columns, descriptions...", then tick up to 10 tables. Golden suggests relationships between them so you can chart across them.
Query Builder: pick a table, columns, and filters, then click Preview Query to see the generated SQL. Copy SQL or Edit in Custom SQL takes it further.
Custom SQL: write a single read-only SELECT. Validate Query checks it before you import. Anything that writes, or more than one statement, is refused with "Only read-only SELECT queries are allowed here".
Click Import Schema to finish.

Tip: Golden remembers the table list for up to 24 hours so the browser opens fast. If a table you just created is missing, click the refresh icon ("Refresh from the warehouse") or the Refresh button next to "Don't see your schema or table?".

The browser lists the datasets in your Project ID, plus a group of Google public datasets for trying things out: usa_names, chicago_taxi_trips, new_york_citibike, san_francisco_311, baseball, github_repos, google_trends, world_bank_intl_education, noaa_gsod, and new_york_taxi_trips.
Data in another project doesn't appear in the browser. Reach it with Custom SQL and a fully qualified name, for example other-project.dataset.table.
Row counts and column details load for the first 20 datasets. Views never show a row count.

How queries and refresh work

BigQuery connections are live. Import Schema saves the shape of your tables, never the rows. Every chart, filter, and AI question runs a fresh query in BigQuery, so what you see is always current.

Golden reuses a chart's result for 5 minutes, so flipping between sheets doesn't re-run the same query.
There's nothing to schedule. Reopen a dashboard and it queries the warehouse again.

Keep an eye on cost

BigQuery bills by data scanned, and Golden doesn't set a cap per query. A few habits keep it in check:

Point Golden at partitioned or clustered tables, and filter on the partition column.
Select the columns you need in Custom SQL instead of SELECT *.
If a chart times out in Golden, the job keeps running in BigQuery until it finishes and is still billed. Narrow the query rather than retrying it.
Set a project-level quota in Google Cloud if you want a hard ceiling.

Browsing tables is cheap: the browser reads table metadata, not the tables.

Limits

Limit	Value
Rows per query	50,000
Chart query time	120 seconds. After that the chart shows a timeout error.
Rows in a table chart	50,000. Totals are still computed over every row.
Queries running at once	4 per dataset. The rest wait their turn.
Query rate	600 per minute per account. Past that: "Too many queries right now. Please wait a moment and try again."
Tables per import	10

Troubleshooting

What you see	What to do
A Reconnect prompt saying your session expired	Click Reconnect on the chart and sign in with Google again.
Reconnect needed, "Ask the connection's owner to reconnect it."	Only the person who set up the connection can reconnect it. Ask them.
A message that your organization's Google security policy requires periodic re-authentication	Your Google admin requires regular sign-in. Reconnect when asked; it's expected.
Invalid Project ID.	Use the project ID, not its display name: 6 to 30 lowercase letters, digits, or dashes.
Failed to list BigQuery datasets	Check the Project ID and that your account has BigQuery Job User and Data Viewer.
A quota or rate-limit error	BigQuery is busy. Wait a moment and try again.