For entry level/junior folks, being able to describe the project and its value to the users is more important than what it actually does. You can build a simple random Pokemon generator and if it attracts 200k users, then that's a good data point to use.

definitely agree with you on this one, "value to user" or "solves a business problem" type projects seem like a great idea where you show an understanding for the a bit of the operation and business side.