Getting Started

Quickstart

From sign-up to your first Slack alert in under 10 minutes.

1

Create your account

Go to app.lighthouseai.co/sign-up and create an account. After signing in, you'll be guided through a short onboarding wizard.

The wizard asks a few questions about your team. These are a one-time survey to help us improve the product — they do not affect permissions or feature access.

FieldWho sees it
DepartmentEveryone
Team sizeEveryone
Your role (job function)Everyone
Company industryOrg creator only
Organization sizeOrg creator only

When the wizard completes, you'll be redirected to /overview — the main dashboard.

2

Connect a warehouse

Navigate to /data-sources and click Create. Select your warehouse type — Snowflake, BigQuery, or Postgres — and fill in the connection credentials.

Important
Always use read-only credentials. Lighthouse never writes to your warehouse. It only reads schema metadata and runs aggregate queries to calculate your metrics.

You must click Test Connection and see a green success state before the save button becomes active.

See Connect a Warehouse for warehouse-specific field requirements and troubleshooting.

3

Create a dataset

Navigate to /data-sets and click Create. Pick the data source you just connected, then choose a table or view.

For each column, assign a role — which columns can be used as metric values, filters, segment dimensions, and time references. At minimum, mark one column as a datetime column so metrics can be scoped to time windows.

Note
As soon as you create a dataset, Lighthouse automatically runs an AI suggestions job in the background. It reads your column structure and proposes metrics worth monitoring. You'll see the suggestions on the dataset detail page.

See Create a Dataset for the full column role reference.

4

Create a metric

Navigate to /metrics and click Create new metric. Choose a Business metric (pick aggregation + column) or write a custom SQL metric.

You can also describe the metric you want in plain English — the Metric Assistant will configure the form for you. Or click Create Metric on any AI suggestion card to pre-fill the form from an AI recommendation.

Before a metric can go live, three sections must be complete:

  • Schedule How often Lighthouse queries your warehouse (e.g. every hour)
  • Threshold The condition that triggers an alert (e.g. value drops more than 30%)
  • Notification channel The Slack channel where alerts are posted

See Creating Metrics for all available options.

5

Connect Slack

Go to your communication sources settings and click Connect Slack. You'll be taken to Slack's OAuth screen — authorize the Lighthouse app to post to your workspace.

Once connected, go back to your metric and select the Slack channel where alerts should be posted.

See Slack Integration for details on the alert message format and interactive buttons.

6

Go live

Once your metric has a schedule, threshold, and notification channel configured, enable it. Lighthouse will begin monitoring on your schedule and post to Slack whenever the threshold is breached.

Tip
The Free plan checks every 4 hours. If you need faster monitoring, upgrade to Professional (1-hour minimum) or Enterprise (5-minute minimum).