I think you always have to consider the effect of what you do. There are many situations - like the ones you describe - where that might be as simple as thinking "huh, this doesn't affect anything" and getting to work. That's still a valuable exercise because it gets you in the habit of at least weighing it.
That seems a bit egotistical.
If I'm trying to make some in-house enterprisey thing run twice as fast, chances are there isn't much in the way of societal impact for me to think about.
If I'm building a sub-sub-component of some new system, chances are I don't have enough information to even start guessing at the overall effects of the complete system as it will eventually end up being used.