JSON to BigQuery Schema Generator Online

Generate BigQuery table schemas from JSON

BigQuery tables are defined by field names, data types, and modes such as NULLABLE or REPEATED. When you have sample JSON payloads from an API, log stream, or ETL job, inferring a schema by hand is slow and error-prone. This tool inspects your JSON object and outputs a BigQuery-compatible field list you can paste into load jobs, Terraform, or the BigQuery console.

What this tool does

Paste a JSON object in the input panel. The generator walks each key, maps JavaScript values to BigQuery types, and nests RECORD fields for objects and arrays. The output is a JSON array of field definitions with name, type, mode, and nested fields where needed.

Type mapping rules

Strings become STRING, with date and timestamp detection for common formats.
Integers map to INTEGER and floating-point numbers to FLOAT.
Booleans map to BOOLEAN.
Nested objects become RECORD fields with child field definitions.
Arrays become REPEATED fields based on the first element type.

Common use cases

Prototype BigQuery tables before loading JSON exports from APIs.
Document expected payload shapes for analytics pipelines.
Generate starter schemas for Cloud Functions or Dataflow ingestion jobs.
Compare inferred schemas against production table definitions during migrations.

Related tools

Build a JSON Schema from sample data with the JSON to JSON Schema Generator, convert JSON to YAML with the JSON to YAML Converter, or format and validate JSON using the JSON Formatter.

Frequently Asked Questions

Does the input need to be a JSON object?

Yes. The root value must be a JSON object because BigQuery table schemas describe named top-level fields. Arrays and primitive values at the root are not supported.

How are arrays handled?

JSON arrays are emitted as REPEATED fields. The tool inspects the first array element to determine the repeated field type. Empty arrays default to a nullable repeated field.

Can I use this output directly in BigQuery?

The generated JSON follows BigQuery field schema structure and can be used as a starting point for table creation or load job configuration. Review inferred types against your full dataset because a single sample may not represent every edge case.

What if my JSON contains mixed types in one field?

BigQuery columns have a single type. The tool infers types from the sample value you provide. If production data mixes types in the same field, coerce or normalize values before loading or choose a broader type manually.

Is my JSON uploaded to a server?

No. Schema inference runs entirely in your browser. Your JSON content is not stored or transmitted to our servers.

Report

JSON To BigQuery Schema