Fifteen real Laravel applications, every one of them a codebase you can go and read. 811 tables. One pinned MySQL 8.0 database, migrations only, no seed data. Every application installed from a clean clone and migrated before a single table was read.
This is not a review of those fifteen projects. It is a census: a look at what a Laravel schema actually looks like once an application has shipped and grown past the tutorial stage. A handful of real issues turned up during the read, and each one is going to the project that owns it, privately, before anything about it is written here. What is left, once those are set aside, is more interesting anyway. These projects share conventions they never discussed, because the framework or the ecosystem around it chose first.
Two different reads sit behind this piece, and they cover different sets, so I keep them apart throughout. The relationship counts cover all fifteen applications. The structural checks are slower, and some of these applications could not run the tooling at the time, so they cover a smaller set: twelve databases and 482 tables, of which seven are real applications and five are bare Laravel baselines kept as a control. Wherever a number appears below, it says which of the two it came from.
