Skip to content
Postgres monitoring

Monitor business KPIs
from your Postgres database.

Lighthouse connects to PostgreSQL with read-only credentials and continuously watches your business metrics. The moment one moves outside its normal range, your team gets a Slack alert, with full context.

Connect Postgres free
  • Free forever tier
  • Read-only access
  • Setup in 10 minutes

Order volume

Postgres
142-47%

Order volume triggered

Value142
Baseline268

Connect in 3 minutes.

Create a read-only Postgres user, paste your credentials into Lighthouse, and you're done.

-- Create a dedicated read-only user for Lighthouse
CREATE USER lighthouse_monitor WITH PASSWORD 'your_secure_password';

-- Grant access to the schemas you want to monitor
GRANT CONNECT ON DATABASE your_db TO lighthouse_monitor;
GRANT USAGE ON SCHEMA public TO lighthouse_monitor;
GRANT SELECT ON ALL TABLES IN SCHEMA public TO lighthouse_monitor;

-- Ensure future tables are also covered
ALTER DEFAULT PRIVILEGES IN SCHEMA public
  GRANT SELECT ON TABLES TO lighthouse_monitor;

Then add your host, port, database name, and credentials in Lighthouse. Done.

Any KPI in your Postgres tables.

Describe what you want to monitor in plain English. Lighthouse generates the SQL.

Transactions

  • Order volume last hour
  • Payment success rate
  • Average order value, daily

Users

  • New signups today vs last week
  • Daily active users
  • Account activations this week

Revenue

  • Revenue last 24h by tier
  • MRR from subscriptions
  • Refund rate this month

Product

  • Feature usage by user segment
  • Session count by platform
  • Funnel step conversion

Retention

  • 7-day retention by cohort
  • Churn rate this month
  • Days since last login, median

Operations

  • Error events in the last hour
  • Job queue depth
  • API response errors logged

Built for real Postgres environments.

Native Postgres connector

Connect with a read-only database user in minutes. Lighthouse reads your schema and is ready to monitor without any manual configuration.

Postgres-optimized SQL generation

Lighthouse generates production-grade Postgres SQL (window functions, CTEs, date_trunc, generate_series time fills) for metrics that work correctly, not just run.

Day-of-week baseline learning

Lighthouse compares each metric to the same day last week and a 4-week rolling average. Monday isn't compared to Sunday. No threshold tuning required.

Rich Slack alerts

Every alert includes current value, baseline, percentage change, threshold, compare period, and segment. Full context delivered to your channel, no dashboard required.

Managed and self-hosted Postgres

Connect to RDS, Aurora Postgres, Supabase, Neon, Railway, or any self-hosted Postgres instance, as long as Lighthouse can reach the host.

No ongoing maintenance

When your schema changes, Lighthouse adapts. No broken cron jobs, no silent query failures, no alerts about the alerting system itself.

Also connects to

SnowflakeBigQueryRedshift

Common questions

What's the easiest way to connect Slack to Postgres?
The DIY route is a pg_cron job or a trigger that runs a query on a schedule, plus a script that formats the result and posts to a Slack webhook, with no baseline or seasonality awareness, so you tune thresholds by hand. Lighthouse replaces that: connect with a read-only database user, describe the metric in plain English, and alerts post straight to Slack with adaptive baselines built in.
What credentials does Lighthouse need for Postgres?
A read-only database user with SELECT permissions on the schemas you want to monitor. We recommend creating a dedicated Lighthouse user; nothing with write access is required.
Does Lighthouse work with RDS, Aurora Postgres, or Supabase?
Yes. Lighthouse connects to any Postgres-compatible host: Amazon RDS, Aurora Postgres, Supabase, Neon, Railway, or self-hosted. You just need to allowlist the Lighthouse IP and provide connection credentials.
Can I monitor application database metrics, not just analytics?
Yes. If the data is in Postgres, Lighthouse can monitor it: user signups, orders, transactions, feature usage, error logs. Many teams use Lighthouse on their production application database as well as their analytics warehouse.
How is this different from Postgres monitoring tools like pganalyze?
Database monitoring tools like pganalyze watch the health of Postgres itself: query performance, index usage, vacuum stats, bloat. Lighthouse monitors the business values inside your tables: revenue, user activity, conversions. These are complementary, not competing.
How long does setup take?
Most teams connect Postgres and create their first metric alert in under 10 minutes. The most time-consuming part is creating the read-only user, which takes about 3 minutes.

Start monitoring your Postgres metrics.

Connect your database, describe a KPI in plain English, and get a Slack alert in 10 minutes. Free to start.

Connect Postgres free

No credit card required · Read-only access · Cancel anytime