Nanocrystalline cellulose (NCC) is a colloidal rigid rod, referred to by various terms in the literature including cellulose whisker (CW) and cellulose nanocrystal (CNC). These charged colloidal rods exhibit complex colloidal phase and rheological behaviours in aqueous suspensions, that are dependent on volume fraction and interparticle forces. A major shortcoming in the literature of NCC is that the dimensions and morphology of NCC particles vary significantly with the type of raw material and manufacturing conditions, which causes inconsistencies in suspension rheology and colloidal behaviours reported between different works. In this review, we consider the theory and experimentally-determined rheological and colloidal phase behaviours of charged rod suspensions in general, with a focus in particular on NCC. Dilute and semi-dilute NCC suspensions are isotropic liquids, in which NCC particles follow diffusional dynamics. The rheology of these isotropic NCC suspensions can be described by theoretical models that account for the effects of rod dimensions and surface charge, including those based on Doi and Edwards' theory. With increasing NCC concentration, the isotropic phase can undergo a transition to a liquid crystalline state (isotropic-nematic transition) or a transition to a dynamically arrested solid (liquid-solid transition). The liquid crystal ordering and gelation/glass transition are of particular interest because they respectively form an ordered structure and allow a solid-like mechanical response at relatively low solids fraction. For conditions at which the isotropic-nematic and liquid-solid transitions coincide, the formation of an anisotropic structure within a soft solid suspension is possible. Investigation of these two competing transitions led to the discovery of liquid crystal re-entrancy and existence of an anisotropic soft solid (liquid crystal hydroglass, LCH). LCH has a biphasic structure with an attractive glass matrix and a co-existing liquid crystal phase, providing similar viscoelastic properties to hydrogels but permitting reversible orientation of the colloidal rods in the liquid crystalline phase by shear forces; i.e. their structural ordering is programmable. The liquid crystal transition and gelation/glass transitions are quantitatively dependent on rod dimensions i.e. respectively proportional to L2D and L/D. Phase transitions in NCC suspensions including liquid crystal re-entrancy and formation of LCH can be fully described as a function of rod dimension, volume fraction and interparticle forces. This behaviour is independent of NCC source, allowing development of a generalised phased diagram in which separately-reported phase transitions converge to consistent phase boundaries. This validates a key hypothesis for the study of NCC suspensions, that variation in NCC concentration and interparticle forces can explain the complex phase behaviours observed within suspensions formulated using NCC obtained from different sources.