> ## Documentation Index
> Fetch the complete documentation index at: https://docs.condense.io/llms.txt
> Use this file to discover all available pages before exploring further.

# BigQuery (Output, Store)

**Connector Type:** Output Connector

**Description**

Stores processed real-time data into BigQuery tables for querying, dashboards, and analytics.

**Brief Description**

The BigQuery Store Output Connector enables real-time event ingestion from Condense Kafka-native topics directly into Google BigQuery tables. This allows users to stream data into Google's cloud-native data warehouse for immediate querying, analytics, dashboards, compliance tracking, and machine learning workflows.

This connector supports continuous write operations from Kafka into a specified BigQuery table using a Google Cloud service account for authentication. It handles streaming inserts, schema matching, and structured JSON payload mapping.

**Links to Relevant Documents**

* [https://cloud.google.com/bigquery/docs/introduction](https://cloud.google.com/bigquery/docs/introduction)
* [https://cloud.google.com/bigquery/streaming-data-into-bigquery](https://cloud.google.com/bigquery/streaming-data-into-bigquery)
* [https://cloud.google.com/bigquery/docs/datasets-intro](https://cloud.google.com/bigquery/docs/datasets-intro)
* [https://cloud.google.com/iam/docs/creating-managing-service-account-keys](https://cloud.google.com/iam/docs/creating-managing-service-account-keys)
* [https://cloud.google.com/bigquery/docs/schemas](https://cloud.google.com/bigquery/docs/schemas)

**Connector Overview**

This output connector continuously consumes structured messages from a Kafka-native Condense topic and inserts them as rows into a specified BigQuery table.

Best suited for:

A. Real-time analytics pipelines.

B. Dashboard integration with Looker, Superset, or Looker Studio.

C. Machine learning feature store population.

D. Regulatory audit trails and compliance reporting.

E. Event archiving with fast SQL access.

**Core Functionality**

1. Streams data into BigQuery using JSON row inserts.
2. Maps event fields to BigQuery table columns.
3. Authenticates using Google Cloud service accounts.
4. Logs schema validation and write errors for troubleshooting.
5. Supports batched inserts for improved performance where applicable.

**How It Works in Condense**

1. The connector subscribes to a Kafka-native Condense topic.
2. Incoming messages are expected to be valid JSON objects.
3. The connector authenticates using the configured Google Cloud Service Account.
4. Each message is inserted into the configured BigQuery table.
5. Insert status and errors are recorded in Condense logs.

**Configuration**

| **Category**   | **Field Name**      | **Description**                             | **Required** |
| :------------- | :------------------ | :------------------------------------------ | :----------: |
| General        | Title               | Unique connector name                       |      Yes     |
| Storage        | Project ID          | Google Cloud project containing the dataset |      Yes     |
| Storage        | Dataset ID          | BigQuery dataset name                       |      Yes     |
| Storage        | Table ID            | Destination BigQuery table                  |      Yes     |
| Authentication | Service Account Key | Google Cloud Service Account JSON key       |      Yes     |
| Condense       | Topic (Input)       | Kafka-native topic to consume from          |      Yes     |

**Field-by-Field Explanation**

**1. Title**

Internal reference name for the connector.

**Examples:**

* bq-fleet-logger
* bq-revenue-stream
* bq-telemetry-archive

**2. Project ID**

The Google Cloud project containing the BigQuery dataset.

The configured Service Account must have access to this project.

**Examples:**

* fleet-intelligence-prod
* customer-analytics-2025

**3. Dataset ID**

The BigQuery dataset containing the destination table.

The dataset must already exist.

**Examples:**

* raw\_events
* vehicle\_telemetry
* streamed\_data

**4. Table ID**

Destination table where streaming records will be inserted.

The table schema should match the incoming JSON payload.

**Examples:**

* driver\_behaviour
* trip\_summaries
* fraud\_alerts

**5. Service Account Key**

JSON credentials for a Google Cloud Service Account.

The Service Account should have the `roles/bigquery.dataEditor` role on the dataset or table.

For security:

* Use dedicated service accounts.
* Rotate credentials periodically.
* Store credentials securely.

The Service Account JSON can be downloaded from:

**Google Cloud Console → IAM & Admin → Service Accounts**

**6. Topic (Input)**

Kafka-native Condense topic from which records are consumed.

Payloads should be valid JSON.

If necessary, use a Condense Transform to flatten or modify the schema before writing to BigQuery.

**Sample Configuration Example**

```json theme={null}
{
  "title": "bq-realtime-vehicle",
  "bigQueryProjectId": "fleet-intelligence-prod",
  "bigQueryDatasetId": "vehicle_events",
  "bigQueryTableId": "geofence_breaches",
  "serviceAccountKey": "{...GCP Service Account JSON...}",
  "inputTopic": "geo-breach-stream"
}
```

**Troubleshooting and Common Issues**

**1. Schema Mismatch**

**Error:** Incoming payload does not match the BigQuery table schema.

**Solution:** Update the table schema or preprocess the payload using a Condense Transform.

**2. Authentication Failed**

**Error:** Invalid Service Account credentials or insufficient permissions.

**Solution:** Verify the Service Account JSON and ensure it has the `roles/bigquery.dataEditor` permission.

**3. Streaming Insert Failures**

**Cause:** Network issues or BigQuery streaming quota limits.

**Solution:** Verify connectivity and monitor Google Cloud quotas.

**4. Payload Parsing Error**

**Cause:** Payload is not valid JSON.

**Solution:** Transform Avro, CSV, or binary payloads into JSON before sending them to the connector.

**Advanced Topics**

**1. Streaming Inserts**

BigQuery streaming inserts provide low-latency data availability for analytics.

**2. Schema Evolution**

Changes to incoming event schemas must also be reflected in the destination BigQuery table.

**3. Partitioning and Clustering**

Use BigQuery table partitioning and clustering for improved query performance on large datasets.

**4. Duplicate Handling**

Streaming inserts do not automatically eliminate duplicate records.

Use unique identifiers or additional Condense logic where idempotent processing is required.

**Best Practices**

1. Keep the Condense message schema aligned with the destination BigQuery schema.
2. Flatten deeply nested JSON where possible to improve query performance.
3. Use descriptive project, dataset, and table names.
4. Monitor BigQuery streaming quotas and connector logs.
5. Store Service Account credentials securely and rotate them regularly.