I'm sorry but that question didn't make sense and I'm not sure what exactly you are looking for.
But to provide an anecdotal point- it's common claimed in programming world that 90% of program's lifetime is contained in 10% of code. Does that then makes the 90% of the code useless? Certainly not. Just having a feature that may be accessed by a program only 1% of time may be considered a big plus by the end users.
Furthermore, another problem with the question is that it's poorly defined. In one application, we may need a lot and lot of this functionality X, in other application, it's not as important. Which should then trump? Does it even makes sense if it was never used for the 2nd application and had no ill effect on the 2nd application's performance?
Yes, we may never use all features in a given product, but to claim that we should then strip them down to the essentials is not always justified. OTOH, if it's feature bloat you're concerned about then in many cases there already are simpler products out there. For example, MySQL is a popular choice for non serious uses because it was simple and easy to use even though the "big boys" had a free edition of their products. There also are many different CAD programs with different design goals, and one would be SketchUp where it's comparatively simple to use. Heck, I always use TextEdit to do my text processing because I don't need all the whizzbang Word has to offer and get in my way.