Here you can see the first sign of trouble. For me as a user (and therefore for your marketing department) this kind of feedback should be the most important one. Your "thing" should absolutely, positively work 99.999% of time. That is why people love Apple so much: their things always work. Nokia has got a much patchier history of things working. Instead, as a developer, you call this kind of feedback "not very helpful" based solely on the fact that you can't address it by simple finger snapping. Yes, I am a developer too, and I know how much we would like to only address feedback that is easy to address. The life is tough though, and this is one of behavioral patterns that will not help you succeed in life. So, you have to constantly fight it, if not as a person, then at least on management level. Your management has to insist that you fix the show-stopping bugs first, not just the "easy" bugs.
This category is different in the sense that it can wait until the next hardware release or at least until the first category is fixed. Nevertheless, if several hundred users scream "give us that damn dpad" at you in unison, do you really think it is a good idea to answer "well, it is not our intended usage case". Why not let your customers decide on the usage case?