for sure PXRD are very useful but you have to combine the obtained information with others techniques, if powder patterns between the two phases are different and there is no solvent (you can determine it by thermogravimetric analysis and DSC which also give you info about the m.p.) and no chemical transformation has taken place (you can check this with IR or NMR spectroscopy) you can think in a new polymorph. The best thing would be to solve the structure by SC or even PXRD. I also think that raman or SS-NMR spectroscopy can be useful.
I fully agree with Dr D'Agostino. Nevertheless, when working with XRPD diagrams, it is not sufficient to obtain the cell parameters from the positions of the diffraction peaks and to compare the intensity of the peaks. First of all, it is worth to perform the decomposition of each peak (in terms of elementary curves) in order to avoid any overlapping with other polymorphs (in small amounts) or amorphous phases. This is very important especially when working with additives during the crystallization. If you are interested in, I'll send you some papers to better explain my remarks.