There is rarely a right solution but there are always wrong ones. Searching for pareto optimal contours is great but implies the existence of pareto pessimal contours where N poor solutions combine to make something maximally rubbish.

Software dev is not in a design space where all things are broadly equivalent with small pros and cons to choose between. It's more a discontinuous surface with orders of magnitude of cost/benefit changes as one meanders over a given cliff.