Welcome to Lighthouse

Lighthouse monitors your business metrics (from Snowflake, BigQuery, or Postgres) and sends Slack alerts the moment something moves. These docs cover everything from initial setup to advanced threshold configuration.

How Lighthouse is organized

Data SourceYour warehouse connection: Snowflake, BigQuery, or Postgres. One per warehouse.

DatasetA table or view within a data source. Defines which columns can be used as metrics, filters, and segments.

MetricA measurement you want to monitor, e.g. "Daily active users, COUNT of user_id." Each metric belongs to one dataset.

ThresholdA rule configured on a metric that sends a Slack or email notification when the value breaches a condition. Thresholds are configured inside the metric, not a separate object.

Data SourceDatasetMetricThreshold

Quickstart

Connect your warehouse, create a metric, and get your first Slack alert in under 10 minutes.

Get started

Connect a warehouse

Snowflake, BigQuery, Postgres, and SQL Server: read-only credentials, connection test required.

Warehouse setup

Create a dataset

Point at a table or view and configure which columns can be metrics, filters, and segments.

Dataset guide

Creating metrics

Business metrics with aggregations and time windows, or custom SQL, with filters, data lag, and schedules.

Metric guide

AI features

Dataset suggestions, metric AI optimization, and the Metric Assistant chat interface.

AI guide

Thresholds & alerts

Absolute, percent-change, and standard deviation thresholds with configurable gating and severity.

Thresholds guide

Slack integration

Connect your workspace and understand what each alert message button does.

Slack guide

Roles & permissions

Org-level admin vs member roles and per-resource access control lists.

Permissions