A friend mentioned a card game today that I hadn't heard of called Boss Monster, and described it as a game where each player is building a dungeon and whoever ends up with the deadliest dungeon wins.
It got me thinking about building dungeons on a budget. Part of the reason One Page Dungeon Contest has had the success it has is that the constraints it imposes encourage a certain amount of creativity. What other constraints might we consider?
The first that springs to mind is literal, in-game budget. Work out prices for 10'x10' squares of cleared area, doors, monsters, traps, etc, and see what people can come up with when their scale is limited by that resource. But that's a very accountanty approach.
A time budget might also be interesting. One Hour Dungeon Contest, anyone? Even if the products themselves end up not being very interesting, I could see such a thing leading to the development of tooling and processes optimized for saving time. Maybe one hour isn't really reasonable - maybe three is enough to get something interesting but still constrained enough for it to matter? I dunno, might take some tuning.