BigQuery Documentation

What it is

BigQuery is Google Cloud's fully managed, serverless analytics data warehouse. You load or stream data into tables, or query files in Cloud Storage directly, and write GoogleSQL; BigQuery handles storage, scaling and compute. The documentation covers loading data, querying, table design, security, BigQuery ML, the bq command-line tool, client libraries and pricing.

Try it for free

The documentation describes two free routes, as of 2026-09:

• The BigQuery sandbox needs no billing account or card. It includes 10 GiB of storage and 1 TiB of query processing a month, but tables, views and partitions expire after 60 days, and streaming, DML statements and the Data Transfer Service are not available.
• The free usage tier, on a billing-enabled project, also includes 1 TiB of query processing a month, without the sandbox's expiry and feature limits. You pay only beyond the free quota.

New Google Cloud customers also get trial credit. Whatever you use, set a budget alert.

The pages worth reading

• Introduction and the quickstarts for the console and the bq tool.
• Loading data: batch loads from Cloud Storage in Parquet, Avro, CSV or JSON, and when to use each.
• Partitioned tables and clustered tables: the two biggest levers on cost and speed. Partition by a date column and filter on it so queries scan less data.
• Pricing: on-demand pricing charges by bytes scanned; capacity pricing reserves compute. Learn to read the bytes-processed estimate before running a query.
• Best practices for controlling costs and optimising query performance.
• External and BigLake tables for querying files in a data lake.

Prerequisites

Good SQL, including window functions. Understanding of why columnar storage makes SELECT * expensive.

Is it worth your time?

Yes, if you chose Google Cloud: BigQuery is at the centre of most Google Cloud data platforms, and it is also used in the free Data Engineering Zoomcamp. Its cost model teaches a lesson every data engineer needs: in a pay-per-scan warehouse, a careless query costs real money.

Common questions

Does SELECT with LIMIT reduce cost? No. In on-demand pricing you pay for the columns scanned, and LIMIT does not reduce the data read. Select only the columns you need and filter on the partition column.

Is BigQuery SQL standard? GoogleSQL is ANSI compliant with extensions such as arrays and structs.