People will often times make mistakes, specifically around development decisions. In a recent article by Brian Moran, he describes how we could make our lives easier by just thinking about easier solutions instead of just digging into processes and trying to make them run faster.
Say you have a business need defined in a way that requires drop-down list box that will be populated with 100,000 rows that users will have to scroll through. You can try to tune the process to make it more efficient at loading 100,000 rows every time the screen is touched. But that tuning does not change the workload.
Rather than simply tuning the existing workload, a better investment in time would be if you helped the designers of that screen understand and accept a solution that doesn’t require loading 100,000 rows. Or perhaps you have a system that makes heavy use of database cursors in a procedural manner of some kind. Most DBA’s know that architecting with a set-based solution will almost always be better than trying to make the cursor-based approach as fast as possible.