Google BigQuery

Svix can deliver webhooks directly to a Google BigQuery table, without your customers having to set up any listener endpoint or write any glue code.

When Advanced Endpoint Types is enabled, your customers will see the option to use a BigQuery destination in the App Portal.

They will be able to configure the connection right in the App Portal:

• projectId — the GCP project that owns the dataset.
• datasetId — the BigQuery dataset that contains the table.
• tableId — the table that receives the rows.
• credentials — a Google Cloud service account credentials JSON object, provided as a string.

Every batch of webhooks received by the endpoint is inserted into the configured BigQuery table.

Destination table

Without a transformation, Svix inserts each webhook into the table identified by projectId, datasetId, and tableId using two columns: id and payload. Svix generates a unique id for each row and writes the raw payload to payload.

The table must already exist before you enable the endpoint. For the default behavior, create it with:

CREATE TABLE `my-gcp-project.my_dataset.events` (
  id STRING,
  payload STRING
);

If you use a transformation (below), you control the columns each row contains, so your table can use any schema you like — as long as it matches the rows your transformation returns.

Transformations

Each webhook is inserted into BigQuery as a row. A transformation returns the rows to insert, where each row is an object whose keys match your table’s columns.

/**
 * @param input - The input object
 * @param input.events - The array of webhooks in the batch. The number of webhooks in the batch is capped by the endpoint's batch size.
 * @param input.events[].payload - The message payload (string or JSON)
 * @param input.events[].eventType - The message event type (string)
 *
 * @returns Object describing the rows to insert.
 * @returns returns.rows - The array of rows to insert. Each row is an object whose keys match the columns of your BigQuery table.
 */
function handler(input) {
    const rows = input.events.map((event) => ({
        id: crypto.randomUUID(),
        payload: JSON.stringify(event.payload)
    }));

    return { rows };
}

input.events is the list of webhooks received by the endpoint, processed in batches.

Each entry in the returned rows array is inserted as a separate row. The object keys must match the column names of your table. To write different columns, adjust the objects in rows and your table schema to match.

For example, if the endpoint receives the following messages:

{
  "eventType": "user.created",
  "payload": "{\"email\": \"joe@enterprise.io\"}"
}

{
  "eventType": "user.login",
  "payload": "{\"id\": 12, \"timestamp\": \"2025-07-21T14:23:17.861Z\"}"
}

The transformation above inserts two rows into your table.