More specifically, understanding the data structures associated with that business and designing the database correctly.
Understanding the concept of database normalization is an important part of designing a database, and it is even more important to understand when
you should "de-normalize" portions of your database. (This is not often recommended, but sometimes the business just doesn't do business in a fully-normalized situation, and that can cause unnecessary overhead in the form of DB I/O in an application.) Usually, however, the problem is that the database is not normalized enough (or not at all....)
Understanding how to design reports based off of the data in your database. Mostly, data is not useful unless it is reported back to someone somehow. Views are incredibly important here (don't de-normalize your database for reporting concerns! Use views instead)
Other concerns involve understanding how your application handles concurrent users and obtaining primary keys; these are more app-specific thatn DB-design specific, though.