• I don't use it for rebuilding indexes or stats.
• I view extract worst performing for optimisation
• I view table growth/ useage of text columns and will look at indexes and sizes on tables
• I view growth stats for databases
• I make use of running processes and the drill down to individual users
• I use the output of the proc cache and stats as well as the memory breakdown
• I look at cpu and io, page life and some of the others.
• I use all the alerts for blocking, deadlocks, cpu, failed jobs, longin running jobs, open trans
• I use the combined alert log for all servers

how's that for a start?