Databuddy

Understand how people use your product: where they come from, what they do, and
where they drop off. Use that insight to decide what to build or improve next.

Databuddy is open-source product analytics for startups. One cookieless script
tracks visitors, custom events, funnels, and goals, with opt-in error and web
vitals tracking. Databunny, the built-in AI analyst, answers questions about
your data and shows the query behind each answer. Feature flags, short links,
and uptime monitoring run in the same dashboard.

• Building a product? Try hosted Databuddy or follow the tracker setup guide.
• Running your own stack? Read self-hosting below. A packaged release is still pending.
• Want to help build Databuddy? Read the contributor guide. Bug reports and docs fixes count too.

Self-hosting

Run Databuddy on your server with Docker Compose. It sets SELFHOST=true, so
events go straight to ClickHouse, while hosted billing and Databuddy's own telemetry are disabled.
Email and AI are optional; see optional services.

Start your instance

These steps need a release with the databuddy-init image. None is published yet;
check releases before starting.
You'll need Git and Docker Compose for this setup.

git clone https://github.com/databuddy-analytics/Databuddy.git
cd Databuddy
git checkout YOUR_RELEASE_TAG
cp selfhost.env.example .env

In .env, set:

• IMAGE_TAG to the release you checked out.
• POSTGRES_PASSWORD, CLICKHOUSE_PASSWORD, and REDIS_PASSWORD to URL-safe passwords.
• BETTER_AUTH_SECRET and DATABUDDY_ENCRYPTION_KEY to separate random secrets.

The template includes local URLs. Compose supplies database connections,
SELFHOST, and browser settings; you don't need to repeat them in .env.

Generate each password and secret separately with openssl rand -hex 32.
Then start Databuddy:

# Start the databases and create their schemas
docker compose -f docker-compose.selfhost.yml run --rm init

# Build the dashboard for your URLs and start the apps
docker compose -f docker-compose.selfhost.yml up -d --build

Open your dashboard URL, create an account, and add your first website.
The stack includes the dashboard, API, Basket event collector, uptime worker,
and short-link service (port 2500). Ports are configurable in docker-compose.selfhost.yml.

For a public instance, replace the template's local URLs with your HTTPS URLs.
Keep the dashboard and API on the same parent domain. Set BETTER_AUTH_COOKIE_DOMAIN, such as .example.com,
to share login across subdomains. Leave it empty for localhost. Rebuild the
dashboard after changing public URLs; they're part of its browser bundle.

No shared parent domain, such as a PaaS default domain like *.up.railway.app?
Leave BETTER_AUTH_COOKIE_DOMAIN empty and set the dashboard's API_PROXY_URL
to the API's internal URL, then rebuild; the dashboard serves the API on its own origin.
On a PaaS, build the dashboard from source with dashboard.Dockerfile; its URLs
are baked in at build time, so there is no prebuilt dashboard image.

Optional services

• Email: For resets, invitations, and alerts, set RESEND_API_KEY and an EMAIL_FROM sender on your verified domain, such as Databuddy <[email protected]>. Leave ALERTS_EMAIL_FROM empty to use the same sender. Once both email settings are configured, set REQUIRE_EMAIL_VERIFICATION=true to require verified accounts. Recreate the services after changes.

• Social login: Set GOOGLE_CLIENT_ID and GOOGLE_CLIENT_SECRET, or GITHUB_CLIENT_ID and GITHUB_CLIENT_SECRET. Register your dashboard's /api/auth/callback/google or /api/auth/callback/github URL with the provider, then recreate the services. Sign-in options appear when configured; password sign-in works without email or social providers.

• Insights: Set AI_GATEWAY_API_KEY in .env, then rerun docker compose -f docker-compose.selfhost.yml up -d --build. Website research also needs CONTEXT_DEV_API_KEY.

• Status pages: Deploy the status app separately with Node, Bun, and dependencies. From the repo root, set your API and status URLs before building:

  export NEXT_PUBLIC_SELFHOST=true
  export NEXT_PUBLIC_API_URL=https://api.example.com
  export NEXT_PUBLIC_STATUS_URL=https://status.example.com
  NODE_ENV=production bun run --cwd apps/status build
  NODE_ENV=production bun run --cwd apps/status start
  

  The app listens on port 3002. Set the matching STATUS_URL in your Compose .env and rebuild the dashboard to enable public links.

• DQL: Requires separate setup: a restricted dql_user and CLICKHOUSE_DQL_URL passed to the API in Compose. Use HTTPS outside loopback and never use the application's admin credentials. See the DQL setup script.

Self-hosting is still evolving. If you get stuck, tell us what happened or ask in Discord.

Upgrade your instance

Back up your databases and .env, check out the new release in the same directory,
and update IMAGE_TAG.
Keep your existing DATABUDDY_ENCRYPTION_KEY so stored data stays readable.
Pull the images, then apply PostgreSQL changes so you can review any prompts:

docker compose -f docker-compose.selfhost.yml pull --ignore-buildable
docker compose -f docker-compose.selfhost.yml pull init
docker compose -f docker-compose.selfhost.yml run --rm init bun run --cwd packages/db db:push

If you decline a change, stop the upgrade. After accepting the changes, create
any missing ClickHouse tables, views, columns, and indexes:

docker compose -f docker-compose.selfhost.yml run --rm init bun --cwd packages/db src/clickhouse/setup.ts

This adds what's missing but never changes or drops existing objects. Columns
and indexes are only added on single-node installs; with CLICKHOUSE_CLUSTER
set, apply them yourself. Apply any extra migrations in the release notes before
starting the updated apps with
docker compose -f docker-compose.selfhost.yml up -d --build.

Stay in touch

Docs · Discord · GitHub issues · Email

Found a security issue? Please follow SECURITY.md.

License

AGPL-3.0, except packages/scan and packages/pulumi, which are MIT and MIT, respectively. Copyright (c) 2025 Databuddy Analytics, Inc.