Abstract: Purpose: This work aims at using atomic pairwise distribution functions (PDF) to analyze residual long range order in amorphous samples prepared using different preparation methods. Methods: Amlodipine besilate (AMB) was used as a model active pharmaceutical ingredient (API). The anhydrate form of AMB was used as received. The hydrate AMB (HY) form was prepared by recrystallization in water. Amorphous AMB (AM) forms were prepared by the following methods: dehydration of HY and milling of HY. The products were analyzed using X-ray powder diffraction (XRPD) and Raman spectroscopy. XRPD data were subjected to PDF analysis using Fourier transformation thereby obtaining residual long range order information. Multivariate data analysis using principal component analysis (PCA) was utilized to interpret the outcome of the PDF analysis. Results: Two different forms of AM were obtained. The amorphousness was verified by a halo present in the XRPD diffractograms. XRPD and Raman data were the same for all AM samples. The PDF analysis did however provide a route for differentiation between the different AM samples. These differences in the PDF analysis were observed for the residual long range order. The diminishing of the long range order was found as follows: milled HY>dehydrated HY. Using PCA on the PDF data samples prepared by the two different methods were seen to group differently. This was most likely caused by the presence of residual long range order of the milled samples. Conclusion: It is shown that there is a difference between the AM samples prepared by milling and dehydration. These differences were not detectable with Raman or XRPD. However, PDF analysis provided a possibility to discriminate between these two preparational routes. PDF analysis may therefore be utilized to obtain a deeper understanding of the amorphous state.