Exploring novel materials with ice recrystallization inhibition (IRI) activity in several fields often starts with a quantitative analysis of ice crystal size change by a splat assay or sandwich assay on a short time scale from 0.5 to 1 h. This study found that this time scale was insufficient to evaluate the IRI activity of cellulose nanocrystals (CNCs) in a model ice cream system-25.0% sucrose solution. No IRI activity was observed in CNCs incubated with ice crystals on a short time scale of 0.5-2.0 h. However, over longer time scales, the growth of ice crystals was entirely inhibited by 1.0% CNCs (between 2 and 24 h) and 0.5% CNCs (between 24 and 72 h) with corresponding final crystal sizes of 25 and 40 μm, respectively. Additionally, ice shaping was observed on a long exposure time, but not on a short exposure time. The findings presented here can be explained by a time-dependent surface coverage of CNCs on ice crystals. The data here indicate the importance of choosing a suitable exposure time for evaluating the IRI activity of new materials and prompt a better understanding of IRI mechanisms involving CNCs.