Here is a step-by-step guide on how you can connect Claude to Kaivo and start analyzing your data with plain language in minutes.

How to Connect Claude to Kaivo

By connecting Claude to your Kaivo-managed BigQuery data warehouse, you can ask questions about your data using plain language directly inside Claude Desktop or claude.ai.

Prerequisites

Before you begin, please note that Claude will have access to all the BigQuery projects your Google account has access to — even the ones not created using Kaivo.

Here’s key details you need in order to connect Claude to Kaivo:

• Client ID

• Client Secret

• BigQuery Project ID

These pieces of information can be easily found in your Kaivo workspace under data tool instructions for Claude.

Step 1: Initialize the BigQuery Connector in Claude

1. Open Claude Desktop or log into claude.ai.

2. Navigate to Settings → Connectors → Browse connectors.

3. Select Google Cloud BigQuery.

4. Click Connect (or click Add to your team if you are an administrator on a Claude Team plan).

Step 2: Enter Your Credentials

1. When prompted by Claude, enter the Client ID and Client Secret provided in your Kaivo workspace under data tool instructions for Claude.

2. Click Continue to install the connector.

Note: This step installs the BigQuery connector to Claude but does not grant access to your specific data yet. If your Claude organization admin has already completed this step for your team, you will automatically skip to Step 3.

Step 3: Authenticate Your Google Account

1. You will be redirected to a secure Google sign-in screen.

2. Sign in using the Google account that is linked to your Kaivo workspace

3. Click Allow to grant Claude access.

Important: Claude will inherit the exact same data access levels as the Google account you sign in with. You can review or revoke this access at any time through your Google Account security settings.

Step 4: Link BigQuery to Claude via Project Description (Recommended)

To avoid typing your BigQuery project ID every time you start a new conversation, we recommend setting up a dedicated Claude Project.

1. In Claude, create a New Project and give it a recognizable name (e.g., Company BigQuery Data).

2. Open the project instructions and add the following text:

   BigQuery project ID: [Insert Your Project ID Here]

3. Save the instructions. Claude will now automatically know how to find your data for every chat created inside this project.

Step 5: Start Querying Your Data

Now you are ready to interact with your data using natural language.

• If you completed Step 4: Open your dedicated project and ask a question directly, such as: "What were our total sales last month?"

• If you skipped Step 4: You must manually include your project ID in your message. For example: "What were our total sales last month? My BigQuery project ID is [Insert Your Project ID Here]"

If you run into any issues during setup, please contact our support team via the chat widget in the bottom right corner.