I agree that "wanting to be good" must be an existing motivation for someone prior to them being able to make utility matter (in fact, I explored this concept quite deeply in regard to virtue in some of my previous research).

I suppose, then, I would say that I concede that your concern is troublesome if we require that ethics not be occurrent. On the other hand, I think this is less problematic if one wishes to say that one's occurrent ethics (those ethics which they actively consider, as opposed to those virtuous traits they have the good fortune to have as habit) require a system for one to follow.

To put it another way.

I am motivated to be good because I have a deep-seeded value for human life or whatever (this is grossly simplified, since it would take a very special kind of mind to have only a single motivation, simply put, I want to be good)

As I go about my day, I don't think about my actions, but still seem to do things in a way that preserves my motivation (I'm nice to the neighbor lady, I work hard, I pay for my lunch rather than skip the check, etc etc).

But then I am faced with a dilemma of some kind in which my motivation, of which I am not particularly aware, must be examined (I must either work hard or spend extra time with my family). It is at this point that I use an ethical system to inform my occurrent morality. In such a case, utility is a perfectly reasonable choice.

The take-away is that no ethical system of any kind is of any value to anyone if they do not want to be good in the first place. But if they do want to be good, then I think it is my obligation as a philosopher to examine methodology that will allow that person to fulfill their motivation in the face of dilemmas.