MCP connectorsBigQuery

BigQuery MCP: connect BigQuery to Claude and ChatGPT

Ask Claude or ChatGPT which campaigns brought in orders last week, what the average basket was per country or how many first-time buyers came back. The AI finds the right dataset in BigQuery, writes the query, runs it and answers in plain words, so nobody needs to book time with an analyst for a simple number.

We build the connector for your company on Google’s own BigQuery API. Each person signs in with their own Google account, so the AI reads the datasets and tables that person has been given in Google Cloud, and it only reads.

Built for your company · Works in Claude, ChatGPT and Copilot

In practice

What your team can do with BigQuery in Claude and ChatGPT

Marketing

Which campaigns paid off

Orders and revenue per campaign, channel or landing page from the web and ad data you already collect in BigQuery, without building a dashboard for each new question.
E-commerce and sales ops

Orders, baskets and returns on demand

Average basket per country, return rate per product group or repeat purchases per month, asked when the question comes up.
Finance

Check a number against the source

Revenue per market or payment method for a period, pulled from the order tables and compared with what the books say.
Data teams

Self-service with the access you already set

People ask through the AI and reach only the datasets they have been granted in Google Cloud. You choose which groups get the connector and which project the queries run in.
Examples

Ask it like this

  • “Which five campaigns brought in the most orders last week, and what was the revenue from each?”

  • “What was the average basket per country in October?”

  • “How many first-time buyers in September ordered again within 30 days?”

  • “Which datasets do we have in the marketing project, and what is in each one?”

  • “Compare return rates for shoes and outerwear this quarter.”

  • “How many sessions came from the Harlow & Finch newsletter last month, and how many of them ended in an order?”

The connector

What we build for BigQuery

Deployed OS has no ready BigQuery connector yet. We build it for your company when you start, on BigQuery’s own API, with the same rules as every other connector: each person’s own login, access decided by group, every call logged.

Datasets
The datasets you have access to in your company’s Google Cloud projects, with their descriptions.
Tables and views
Tables and views with their columns, types and descriptions, so the AI picks the right one.
Read-only queries
One query at a time that only reads, with a limit on how many rows come back and on how much data a single question may scan.
Your definitions
What counts as an order, a returning customer or a campaign at your company, mapped to the right tables and columns.
Security

Who sees what

Read-only
The connector browses and runs queries that read. Statements that would create, change or delete tables or rows are refused. We only build write actions if you ask for them.
Each person’s own Google access
Everyone signs in with their own Google account. Google Cloud checks every query against the access that person has been given, so a dataset marketing cannot open in BigQuery stays closed through the AI too.
Your company decides who gets it
Admins choose which groups get BigQuery in Deployed OS. A group without access never sees the connector.
Every call logged
Who asked, which query was run and when. Deployed OS keeps the log, and it can be exported.
Setup

Set up the BigQuery connector in three steps

01
We go through with you which questions matter most and which datasets answer them, and build the connector on the BigQuery API.
02
Your Google Cloud administrator approves the connection, and you choose which groups get BigQuery in Deployed OS.
03
Each person connects with their own Google account. From then on they just ask.
Compared

BigQuery’s own MCP server or Deployed OS?

Google has its own BigQuery remote MCP server. It is hosted by Google, comes with the BigQuery API, and uses Google sign-in together with the access you have set in Google Cloud. If BigQuery is the only system your team wants in its AI tool and the people asking are comfortable with Google Cloud, it is a good place to start.

Deployed OS is for when BigQuery is one system among several. It sits next to your CRM, ad accounts and other tools in one company connection, admins decide which groups get each system, every call is logged in one place, and the same rules apply in Claude and ChatGPT. We also build the tools around your own definitions, so marketing and finance can ask without knowing the table names.

Their documentation: BigQuery remote MCP server

Questions

BigQuery MCP: common questions

Yes. Google offers the BigQuery remote MCP server, hosted by Google and governed by the access you set in Google Cloud. Deployed OS adds BigQuery to one company connection with your other systems, with access per group, one log and tools built for your company.

No. The connector lists datasets and tables and runs queries that only read. It does not create, change or delete anything, and we only build write actions if you ask for them.

Yes. Everyone signs in with their own Google account, and Google Cloud checks every query against the access that person has. The AI cannot read a table the person could not query in BigQuery.

We build the connector with a limit on how much data a single query may scan, set together with your data team. A question that would go over the limit is stopped before it runs.

Deployed is an EU company and fully GDPR compliant, and you choose where your data is stored, down to the city. Query results are fetched from BigQuery when someone asks for them.

It depends on the size of your team. Get in touch and we will walk you through it.

More connectorsAll connectors
Let’s talk

See it on your own tools.

Tell us which systems your company runs on. We’ll show you what your team can ask, and what IT gets to control.