Crystallography loves order, but many organic materials are disordered and only partially, if at all, crystalline. Nonetheless, these disordered materials are functionally complex and require characterisation. Their lack of crystallinity poses not only fundamental questions about how to best describe such structures, but also blunts the typically precise tools of crystallography in describing atomic order. Yet, disordered structures not only have structural characteristics, but also complex micro- and nanostructures, defects, and phase distributions. Much of this can be gleaned from diffraction spots, but even more of the information lies in-between the spots in a diffraction pattern. The diffracted intensity outside of the diffracted spots contains the necessary information to not only obtain structural information, but to also characterise the disorder present. Recent developments in transmission electron microscopy (TEM) have enabled the collection of numerous spatially separated diffraction patterns across a specimen, and when combined with computational tools opened a new space for the crystallographic analysis of disordered materials. In scanning electron diffraction (SED), a two-dimensional diffraction pattern is acquired at each probe position in a two-dimensional scan across a specimen. This four-dimensional (4D) data set can be extensively manipulated post-acquisition using computational tools, enabling the acquisition of multiple correlated conventional TEM experiments at once. Yet this is just the tip of the iceberg. Within such a 4D data set, any pixel can be correlated with another, even ones that may seem at first glance unphysical. In this work, such computational microscopy is applied to SED data to characterise the structure of disordered materials. In this work, the requisite computational methods are developed and applied to extract crystallographic information in metal-organic frameworks through pair distribution function analysis, in pharmaceutical cocrystals through nanoscale twist characterisation, and in polymers through semi-crystalline variance and correlation analyses. As all information is contained within a single scan, all of this analysis is done at doses low enough to avoid irradiation damage in the probed beam-sensitive samples.