I try to leave decisions reversible
Picking the simpler option is easy. The useful part is writing down what would have to change if the constraint changes, right next to the code it affects. That turns a future migration into a small edit instead of a rewrite.
Picking SQLite over Postgres mattered less than making that choice reversible. I left the exact swap point documented in the code instead of treating the database as permanent. The payment flow needs the same treatment next: proof-of-payment review works fine at one-organiser, weekly-event volume, but it won't scale past that without real payment integration.