Database

Managed database services run, back up and scale your database for you — so you get a connection string instead of a server to administer.

Managed database services hand you a production database without the operational burden of running one. The provider handles provisioning, patching, backups and scaling; you get a connection string. For most projects that is the right call — administering your own database is a specialized job, and the free tiers here are enough to develop and even launch small applications.

The modern twist in this category is serverless, branchable Postgres. Platforms like Neon separate storage from compute so the database can scale to zero when idle — which keeps free tiers sustainable — and can spin up instant branches that are full copies of your data. Branching changes how you work: every pull request or migration test can run against its own isolated database, something traditional managed offerings simply don’t provide. The trade-off for scale-to-zero is a brief wake-up on the first query after idling.

When comparing free tiers, weigh storage size and any row limits first, then compute and connection limits (connection pooling matters a lot for serverless app patterns), then the workflow features like branching, backups and point-in-time recovery. If you also need authentication and file storage, consider whether a BaaS that bundles Postgres would serve you better than a standalone database — and if you only want the database, these focused services keep things lean.

Free Database platforms

FAQ

What does 'scale to zero' mean for a database?

Serverless databases like Neon can pause compute when idle and resume on the next query. That makes the free tier cheap and sustainable, but the first query after a pause incurs a short wake-up delay — fine for dev and low-traffic apps, worth noting for latency-sensitive ones.

Why would I use database branching?

Branching gives you an instant, isolated copy of your database — ideal for testing a migration, running CI against real schema, or giving each pull request its own data. It's a workflow that traditional managed databases don't offer.

Do I need a managed database if my BaaS already has one?

Often not — a BaaS like Supabase already includes Postgres. A standalone managed database makes sense when you want just the database (no auth/storage layer), need a specific engine, or are connecting it to your own backend.

Other categories