The purpose of computing is insight, not numbers.
This famous quote appeared in the preface of a 1962 book on numerical methods by Richard Hamming (inventor of Hamming code, Turing award winner, and a past president of the Association for Computing Machinery). Scan&Solve™ computes numerical approximation to an assumed mathematical model of linear elasticity. The results of the computation are expressed in terms of numbers that are generally useful in gaining insights into structural properties of solids, but should not be confused with accurate representation of physical reality for many reasons explained in the section on known limitations.
The best way to gain an insight into structural properties of a solid is to look at how the predicted stresses and deformation change with changes in the solid model, material properties, and boundary conditions. During the design process, frequent and consistent use of Scan&Solve™ can help identify poor/great design decisions; choose optimal sizes, dimensions, and shapes; and test the model under a variety of operating conditions and "what if" scenarios. But it is also important to remember that critical design decisions require experimental data and substantial experience; they should never be made based solely on a software simulation. Simulation is not intended to replace physical testing of prototypes, which is required to validate any design.
Analyze early, analyze often ... in Rhino.