Why spend 50% development for something that is used for 5%?

This topic comes up for every project I have worked on. What I am talking about is the dreaded Options/Preferences functionality. I have no scientific analysis to back this up, but I always spend more time working and getting the functionality of the Options/Preferences portion of a piece of software working correctly than any other section. Why is that? Why spend 50% of your time developing something that the average user will spend less than 5% of the time looking at? I think we as developers are made to develop the Options/Preferences functionality so flexible that the complexity is increased. I have found that a good indicator of how well your database is designed may be how complex your Options/Preferences functionality is. That would be a good topic of discussion for a group of software architects IMHO. How to keep in mind when developing a good database and architecture framework to allow the user preferences to just drop out of the design may be a difficult thing to handle. Well that is my thoughts for this morning. If you have anything to add please do.