Given a complex problem, if there is just one part of it that I fear may not be solvable, then I will look into that part first to figure out if a solution exis

Given a complex problem, if there is just one part of it that I fear may not be solvable, then I will look into that part first to figure out if a solution exists and only later into the remaining elements of the problem. If instead the uncertainty comes just from the "size" of the problem, then I split it into smaller parts that are easier to face, and I look into each of them the bare minimum to be able to confirm that a solution exists. Once I am confident a solution exists for all of the smaller parts and therefore for the whole problem, then I start the actual work.