I think that there are two (at least) kinds of work. One kind involves operating within a fixed set of rules. You take a bunch of stuff as fixed and then try to solve problems. There's a certain kind of comfort associated with this work. You really make progress. You can clearly see that stuff gets done. Its also comforting because you don't have to constantly think about the big picture. Focus on the details. Or maybe you find it uncomfortable, because you like to question assumptions and don't like taking things for granted; matter of personal style I suppose.
The other type of work involves trying to find the best overall framework to use in going about solving problems. To do this work, you will probably do similar work as the detailed work I just described. The difference is that you constantly relate it to the big picture. You constantly ask how this problem would have been formulated differently had the overall framework been different. Its hard to know whether anything gets done in this type of work. Its always a moving target. A certain result may stay the same, but seen in a different context takes on a much greater or lesser importance. The difficulty of this kind of work is that its really hard to move forward. Any step involves a million questions about its relationship to the big picture.
I think I am good at the second kind of work and not so good at the first. I'm not very good at just sticking with a set of rules and working within that framework. It makes me slow. Instead of going ahead to the end of a project, I constantly question the framework under which the project was given. I'd like to be able to turn off this constant context checking sometimes. it'd be easier to get work done.
I suppose this perspective is similar to Kuhn's "normal" vs. "revolutionary" science, but applied to all sorts of things, not just science.