Why projects keep landing on Postgres

PostgreSQL has become the default choice for new projects, and the reasons are practical rather than fashionable.

It is genuinely free and open, with no licensing surprise at scale and no vendor deciding your terms.

It is strict about correctness. It enforces constraints properly, it does not silently coerce data into a type it was not given, and it does not quietly truncate a value that does not fit. MySQL historically permitted several of these, and a database that accepts wrong data quietly is a database that will surprise you later.

Its type system is unusually rich. Proper date and time types with time zone handling, arrays, ranges, and JSON stored in a binary form that can be indexed and queried. That last one matters: you can keep structured columns for what you know and a JSON column for what varies, without running a second database.

Full text search is built in, which is adequate for most applications and removes a whole external service.

Extensions add substantial capability, including geographic data and vector search for AI applications, which is why so much recent AI infrastructure is built on it.

It handles concurrent readers and writers well, because readers do not block writers and writers do not block readers.

Where MySQL still wins. Simpler replication, and being what most shared hosting offers, which matters for small PHP applications in this market.