Golden works where Google works.
Gemini for AI reasoning, BigQuery queried live, Google Sheets as a data source, and sharing via Google Slides and Google Sheets
One workflow across the Google stack
BigQuery is queried live
Golden generates SQL and sends it to BigQuery, then renders the result. BigQuery does the work and your project keeps the record. There is no extract to refresh and no copied dataset drifting out of date.
Google identity decides access
People connect with their Google account, so the IAM roles and dataset permissions your admins already manage decide what comes back. Golden does not ask your team to maintain a second permission model beside the one Google enforces.
Answers land in the tools people already use
A question can start in a spreadsheet and end in a deck. Golden reads Google Sheets as a live source and sends finished analysis to Google Slides. Nobody has to leave the tools they already use.
Four places Golden meets Google
Gemini
Gemini is one of the models Golden can use for reasoning: reading a question, proposing an analysis, and explaining a result.
BigQuery
Golden queries BigQuery live. It generates SQL, BigQuery executes it under the connected Google identity, and Golden renders the result. No extract, no duplicate dataset, and no second permission model to maintain.
Google Sheets
A sheet is a real data source, not a one-time upload. Golden reads it in place, so a forecast or budget your team keeps in Sheets can sit alongside warehouse data in the same analysis or dashboard.
Google Slides
Finished analysis exports to Google Slides, so the chart or story that answered the question becomes the deck you present from. The narrative goes with it rather than being rebuilt slide by slide.
How Golden talks to BigQuery
| Step | What happens |
|---|---|
| 1 · Connect | Connect with a Google account or a service account your admin approves. Golden never self-provisions access. |
| 2 · Prepare | Clean, normalize, and reshape data in Golden. The steps compile into SQL that runs in BigQuery, so no data is copied out. |
| 3 · Query | Golden generates SQL and sends it to BigQuery. No extract or duplicate dataset is required. |
| 4 · Enforce | BigQuery executes the query under the connected identity, so IAM roles and dataset permissions apply. |
| 5 · Answer | Golden renders the result for analysis and dashboards, and can send it on to Google Slides for presentations or Google Sheets if you just want the data. |
Why Golden on Google is different
| Criterion | Traditional BI | Golden Analytics |
|---|---|---|
| BigQuery access | Often relies on extracts, copies, or refresh jobs | Queries BigQuery live. No copied datasets |
| Identity | Often uses a shared service account plus app-level permissions | Connects as the user. Google IAM decides access |
| Spreadsheet data | A sheet is a one-time upload that goes stale | Google Sheets is read in place as a live source |
| Delivery surface | Findings are rebuilt by hand in a slide deck | Analysis exports to Google Slides with its narrative intact |
| AI experience | Chat panel separate from the workflow | AI works inside the same workflow users edit and inspect |
| Time to first insight | Training and setup slow teams down | Start with a question, chart, dashboard, or story in minutes |
| Full workflow | Prep, analysis, and storytelling often live in different tools | Connect, discover, prepare, analyze, and communicate in one place |
Golden connects to the rest of your stack too
Snowflake
AI analytics on the Snowflake data, roles, and metrics you already run the business on. Reads your Semantic Views; OAuth means Snowflake roles decide access.
Databricks
AI analytics on the lakehouse data, Unity Catalog grants, and metrics you already run the business on. Every statement lands in your Query History.
Alteryx
Alteryx prepares the data. Golden makes it sing. Curated data that took hours to prepare, in front of every analyst and business user in minutes.
Golden × Google
How does Golden connect to BigQuery?
Golden runs on top of your BigQuery project and queries it live. When someone asks a question or builds an analysis, Golden generates SQL, sends it to BigQuery, and BigQuery executes it; Golden then renders the result for analysis, dashboards, and stories. There are no extract jobs, no second warehouse, and no copied dataset, so every answer comes from live data in BigQuery.
Does Golden move or copy data out of BigQuery?
No. Golden does not extract, copy, or sync your BigQuery data. The data stays in your project, and Golden returns only the results needed for a given analysis, dashboard, or story. There is no duplicate dataset to maintain and no stale copy that drifts out of date, so what people see reflects BigQuery at the moment they ask.
How does Golden respect Google IAM permissions and governance?
Golden passes the connected Google identity through to BigQuery, so Google governs access. IAM roles, dataset and table permissions, and any row-level or column-level policies your admins configured decide what a person can see, because BigQuery executes the query rather than a separate engine. Golden does not maintain a second permission model that could drift away from the one your team already manages.
How does Golden use Gemini, and what leaves my environment?
Gemini is one of the models Golden can use for reasoning: interpreting a question, proposing an analysis, and explaining a result. The model works with the question and the schema it needs, not a bulk copy of your tables, and the analysis it produces stays inspectable and editable in Golden. If your team has a preferred model, or wants to review exactly what is sent, ask us before you connect and we will walk you through it.
Can Golden read a Google Sheet as a live data source?
Yes. Golden reads Google Sheets in place, so a sheet behaves like any other source rather than a one-time upload. That matters for the numbers teams keep in spreadsheets, like targets and forecasts. They can sit in the same analysis or dashboard as warehouse data, and they stay current as the sheet is edited.
Can Golden send analysis to Google Slides?
Yes. Finished analysis can be exported to Google Slides, so the charts and the narrative that answered the question become the deck you present from. That skips the rebuild. Instead of screenshotting charts into a new deck by hand, the story travels from the analysis into Slides, where colleagues can review and comment on it.
What authentication does Golden support for Google?
Golden uses Google OAuth so people connect as themselves, and supports service-account credentials where a team prefers them. Your admin approves the connection and the scopes it requests, and Golden never self-provisions access. Credentials and tokens are encrypted at rest with per-account KMS envelope encryption, and access can be revoked on your side at any time.
Can Golden use Google sources alongside my other data?
Yes. In addition to BigQuery and Google Sheets, Golden connects to Snowflake, Databricks, Redshift, and files such as CSV, Parquet, and JSON. Teams can run one AI analytics workflow across all of these sources, with warehouse queries running in place rather than being duplicated into Golden. A single dashboard can draw on BigQuery and a sheet without either being copied first.
Bring Golden to your Google stack
Connect BigQuery and Sheets, keep Google identity in control, and take the answer straight into Slides.