Supabase as Postgres wrapper
Founding engineer at healthtech startup on Supabase's ready-at-scale credibility gap
Supabase can charge well above raw Postgres pricing, but its ceiling is set by how much of the product feels like saved setup work rather than irreplaceable infrastructure. For experienced teams, auth, row level security, vector search, and API generation often look like conveniences built on familiar Postgres primitives. That makes Supabase strongest when speed, bundling, and operational simplicity matter more than deep technical differentiation.
-
The split is visible in user behavior. A healthtech founding engineer saw little reason to pay for features their team could build on Aurora Postgres, including row level security and newer AI features. A public sector startup, by contrast, uses Supabase for core data, auth, and storage because the bundle works and migration would take months.
-
PG Vector helps product completeness more than moat. In the interview, vector support is described as useful because it is already there, not because it is hard to reproduce. That matches the broader market shift where pgvector narrowed the gap between dedicated vector databases and standard Postgres stacks.
-
The real pricing leverage comes from owning more of the application stack. Supabase monetizes database compute, storage, bandwidth, and auth usage, and the auth layer adds sticky user based pricing. That is enough to build a large business, but it is a different kind of leverage from platforms like Datadog or Snowflake that sell hard to replicate performance and analytics infrastructure.
Going forward, Supabase is likely to keep winning where the buyer wants one place to stand up a backend fast, especially in AI app creation. To push further into enterprise pricing, it will need more products that feel less like polished Postgres packaging and more like difficult infrastructure a serious team would not want to rebuild.