Everything you need to get started

Getting Started

Golden Analytics connects to your data warehouse with read-only credentials, defines metrics against your existing tables, and runs automated attribution on any movement you want to investigate.

Quick start

Follow these steps to connect your warehouse and run your first attribution query.

Connect your warehouse

Golden Analytics supports Snowflake, Google BigQuery, Amazon Redshift, and Databricks SQL. Each connection requires read-only credentials scoped to the schemas you want to analyze.

Snowflake

Create a dedicated service account with read-only privileges on the relevant schemas. Golden Analytics uses the Snowflake JDBC driver and connects over TLS 1.2+. Add the Golden Analytics IP range to your allowed IPs list if your Snowflake account uses IP allowlisting.

-- Create a read-only role for Golden Analytics
CREATE ROLE golden_analytics_ro;
GRANT USAGE ON DATABASE analytics_db TO ROLE golden_analytics_ro;
GRANT USAGE ON SCHEMA analytics_db.public TO ROLE golden_analytics_ro;
GRANT SELECT ON ALL TABLES IN SCHEMA analytics_db.public
  TO ROLE golden_analytics_ro;

CREATE USER golden_analytics_svc
  PASSWORD = '<strong_password>'
  DEFAULT_ROLE = golden_analytics_ro;
GRANT ROLE golden_analytics_ro TO USER golden_analytics_svc;

BigQuery

Create a service account in Google Cloud IAM and grant it the bigquery.dataViewer role on the relevant datasets. Download the service account JSON key and paste its contents into the BigQuery connection form.

Redshift

Create a user with SELECT-only grants on the schemas you want to use. Ensure the Redshift cluster's security group allows inbound connections from the Golden Analytics IP range on port 5439.

Define your first metric

A metric in Golden Analytics is a named aggregate over a table, with associated dimension columns. Once defined, it can be queried, attributed, and scheduled.

Metric definition structure

{
  "name": "Monthly Recurring Revenue",
  "table": "analytics_db.public.subscription_events",
  "value_column": "mrr_usd",
  "aggregation": "sum",
  "date_column": "event_date",
  "dimensions": [
    "customer_tier",
    "vertical",
    "acquisition_channel",
    "geo_region"
  ]
}

Supported aggregations

Attribution model

Golden Analytics uses a variance decomposition approach to attribute metric changes to their contributing dimensions. When your MRR drops by $80k between two periods, the attribution engine calculates how much of that movement each segment combination (enterprise + APAC, SMB + churn-cohort) contributed.

How it works

For each dimension in your metric definition, the engine evaluates the expected contribution under a counterfactual scenario (what would have happened if only this segment changed?). The residual across all segments accounts for joint effects and interaction terms.

Significance threshold

The engine only surfaces segments whose contribution exceeds a configurable significance threshold. The default is 3% of total movement. You can lower this threshold for finer-grained analysis or raise it to filter out noise in high-cardinality segment spaces.

API overview

The Golden Analytics REST API lets you query metrics, retrieve attribution results, and manage stakeholder report configurations programmatically. All endpoints use HTTPS and return JSON.

https://api.goldenanaltyics.com/v1

Authentication

Pass your API key in the Authorization header as a Bearer token.

curl -H "Authorization: Bearer gldn_live_xxxxxxxxxxxxxxxx" \
  https://api.goldenanaltyics.com/v1/metrics

Query endpoint

Submit a natural language query against a defined metric.

{
  "metric": "monthly_recurring_revenue",
  "question": "Why did MRR drop in March compared to February?",
  "date_range": {
    "start": "2026-02-01",
    "end": "2026-03-31"
  }
}

Response includes the attribution breakdown, ranked segment contributors, and a plain-language summary of the top movers.

Metrics endpoint

List all metrics defined in your workspace.

[
  {
    "id": "monthly_recurring_revenue",
    "name": "Monthly Recurring Revenue",
    "aggregation": "sum",
    "dimension_count": 4,
    "last_run": "2026-07-14T09:12:33Z"
  }
]