Importing from raw data or other databases. Exporting to data warehouses.
Cleaning legacy data (Especially unnormalized data. See some inspirational material at Database normalization the easier way. Also, columns with multiple values, such as name and address in one field).
Using DBI with cron jobs. Security implications related to the cron user grants. How to test for a cron job with DBI (it works on the command line but fails at 2:00AM).
On the same vein, using the DBI in CGI or mod_perl scripts can lead to problems related to user authorizations. Many people here in the Monastery ask why their script runs fine interactively and fails when Apache takes over. (see Protecting your DBI user/password in scripts?)
I guess that data and schema movement also implies finding differences between databases and synchronizing them (see DBI -- Synchronize table diffs?).
It might be OT, but you may find some ideas from the MySQL Cookbook table of contents.